of transportation - Главнаяlib.madi.ru/fel/fel1/fel12p004.pdf · international cooperation...

SCIENCE JOURNAL OF

TRANSPORTATION

Especial Issue No. 04 International cooperation Journals

MADI - SWJTU - UTC

Moscow - Chengdu - Hanoi 01 - 2012

Dear researchers, colleagues and readers, Transportation is the means by which all people are connected, all human activities occur. Nowadays, in strong globalizing process, community activities have not been limited by countries’ borders; thus transportation becomes non-confrontiers. We, transportation makers, in this moment, have had a common forum to together discuss, contribute, share and dedicate. First Especial Issue of international co-operating transportation science journals of State Technical University (MADI)-Russia, South West Transportation University (SWJTU)-China and University of Transportation and Communication (UTC)-Vietnam is published in spring – season of blooming and developments. We wish you and our transportation career were achieved, prosperous and fruitful. Science is non-limitation, Transportation is non-border, Friendship is non-confrontiers, Aim toward the future, we will do our best to make transportation: More intelligent and effective, Faster and safer, Cleaner and greener, With that objective, by this forum, we together connect, endeavour, research, create, contribute, share and devote. Moscow-Chengdu-Hanoi Board of Editors-in-Chief

ISSN 2410-9088

SCIENCE JOURNAL

OF

TRANSPORTATION

Especial Issue No. 04

International cooperation Journals

State Technical University-MADI Southwest Jiaotong University (SWJTU)

University of Transport and Communication (UTC)

Moscow - Chengdu - Hanoi 06-2012

TABLE OF CONTENT

Pages 3 PROF. VIKTOR A. KORCHAGIN LGTU, Lipetsk, Russia

PROF. VALENTIN V. SILIANOV MADI, Moscow, Russia

ASSOC. PROF. YULIYA N. RIZAEVA LGTU, Lipetsk, Russia

Dynamic adaptive control by advancement of traffics of goods

Pages 11 TRAN TUAN HIEP TRAN VU TUAN PHAN

Civil Engineering Faculty, UTC

Sustainable developing road transport system for Hanoi

Pages 18 QI WANG HAILI LIAO

MINGSHUI LI CUNMING MA

Research Center for Wind Engineering, Southwest Jiaotong University,

Chengdu 610031, China

Influence of aerodynamic configuration of a streamline box girder on bridge flutter and vortex-induced vibration

Pages 29 KRASNYNSKIY M.N. NIKOLAEV A.B.

OSTROUKH A.V. State Technical University - MADI,

Moscow, Russia

Application of virtual simulators for training students in the field of chemical engineering and professional improvement of petrochemical enterprises personnel

Pages 33 QIHENG LU XIAOYUN FENG

Southwest Jiaotong University, Chengdu 610031, China

Optimal control strategy for energy saving in trains under the four-aspect fixed autoblock system

Pages 42 PROF. DR. NGUYEN VIET TRUNG DR. TRAN VIET HUNG

University of Transport and Communications, Vietnam

Y the oriented application of weathering steel for bridge in Vietnam Pages 52 TIAN LI

JIYE ZHANG WEIHUA ZHANG

Traction Power State Key Laboratory, Southwest Jiaotong University,

Chengdu 610031, China An improved algorithm for fluid-structure interaction of high-speed trains under crosswind Pages 61 BOGDANOV N.K.

ZAMYTSKIKHY P.V. KHADEEV A.S.

State Technical University - MADI, Moscow, Russia

New approaches to the choice of architecture for a supervisory system of gaz transportation Pages 65 TRAN XUAN TRUONG

Telecommunication Engineering Department, UTC

Improvement quality of congestion controller in ATM network by method using neural network Pages 72 MSC. VU KIM HUNG

Institute of Transport Planning and Management University of Transport and Communications

environmental indicators for sustanable development urban transport planning in Vietnam Pages 78 BARINOV K. A.

KRASNYANSKIY M. N. MALAMUT A. J.

OSTROUKH A. V. State Technical University – MADI,

Moscow, Russia Algorithm of virtual training complex designing for personnel retraining on petrochemical enterprise

INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04 1

Pages 82 XIAOYU GUAN YONG ZHAO XIAOQIU JIA


Ni doping effects YBA2FE3O8 + W Pages 89 PH.D. KURENKOV P.V

Department Transport business MIIT GRADUATE POLYAEVA T.I

Department Economics and Logistics in the transport SamGUPS

Economic competitive evaluation of high - speed lines Pages 96 PhD. VU THE SON

University of Transport and Communications

MA. NGUYEN VAN SUU Transport Engineering Design

Joint Stock Incorporated South MA. TRAN HUU BANG

Sai Gon Construction Quality Control Joint Stock Company

Calculation process of settlement - transition pile net for bridge approach embankments Pages 104 XUEYI LIU

PINGRUI ZHAO FENG DAI


Advances in design theories of high-speed railway ballastless tracks Pages 117 MAI VINH DU

DUOQIAN MIAO RUIZHI WANG

Department of Computer Science and Technology Tongji University,

Shanghai 201804, China LE HUNG LAN

Electrical & Electronic Department, The University of Transport

and Communications, Hanoi, Vietnam An efficient method for Vietnam license plate location

Pages 127 YINGXUE WANG BO GAO

CHAO ZHANG XUZHOU HE

School of Civil engineering, Southwest Jiaotong University,

Chengdu 610031, China Analysis on setting airshaft at mid-tunnel to reduce transient pressure variation Pages 134 TRAN THE TRUYEN

Department of Civil Engineering University of Transport

and Communications Service life estimation of reinforced concrete structures with considering the damage of concrete cover Pages 142 BERNER, L.I.

KOVALEV A.A. NIKOLAEV A.B.

State Technical University – MADI, Moscow, Russia

The block of modeling and forecasting of gas networks systems operation modes for dispatcher decision support system Pages 146 GALINA BUBNOVA

Moscow State University of Railway Transport

Russia Industrial integration and economical crisis

Pages 156 DR. LE VAN BACH University of Transpor

and Communications ENG. TRAN HUU BANG

Saigon Construction Quality Control Jont Stock Company

Initial result of using cinder particles at some steel factories in Ba Ria - Vung Tau province as a mineral additive for cement concrete in building motoray sufrace Pages 163 DR. ALEXANDER CHUBUKOV

PROF. VALENTIN SILYANOV State Technical University - MADI

Mechanisms for integration federal and regional strategy for ensuring road traffic safety in Russia


DYNAMIC ADAPTIVE CONTROL BY ADVANCEMENT OF TRAFFICS OF GOODS

PROF. VIKTOR A. KORCHAGIN LGTU, Lipetsk, Russia PROF. VALENTIN V. SILIANOV MADI, Moscow, Russia ASSOC. PROF. YULIYA N. RIZAEVA LGTU, Lipetsk, Russia

Summary: On all stages of functioning of transport systems approach of the systems is used to the choice of rational management exploitation of motor transport. Intellectual control system is offered transport streams in cities on the basis of traffic-light objects for reduction of expenses of resources of by participants process of delivery of loads.

Key words: A transport system, transport stream, city.

Transport possibilities of Russia, along with natural resources and geographical location, - the competitive edge of our country. An increase of their attractiveness is a task state, as often nascent transport corks and delays of people and loads on motorways with plenty of the lost and trauma people, height of consumption of power resources and negative influence on an environment in world practice it is accepted to characterize as strategic problems of national level.

CT 2

Presently for many key objects of transport systems of our country uncoordinated and coordinated not enough co-operation is very characteristic in-process motor transport. It, foremost, contingently inharmonious development of associate infrastructural links of transport, стыкующихся in knots, absence for them of single informative and normatively-legal base, management center. As a result in transport knots and on going failures, congestions in advancement of transport streams, scale outages of rolling stock, appear near them, the terms of delivery of loads increase, both in internal and in international reports.

Deciding these tasks is possible on the basis of application of modern methods of organization of motion of transport vehicles and pedestrians, by development and introduction of intellectual control system (ICS) by transport streams of city. Without application of computer facilities to decide this problem practically impossible.

Unsolved tasks in organizations of travelling motion, considerable increase of intensity of a transport stream create serious problems in organization of transportations, worsen efficiency of


transport processes. One of variants of increase of efficiency of a transport process is a dynamic management by travelling motion. A dynamic management has a row of advantages motion as compared to a local management, because this system has the opportunity to predict non-standard situations, forecast the changes of basic parameters of a transport stream, co-ordinate work of technical equipments of adjusting, and also to coordinate managing parameters on a network. It is important in transport knots to provide not only co-ordination in-process TCS but also co-operating with them dispatch, insurance companies, broker-custom organs, freight terminals, storages of temporal storage, operators of rolling stock and other partners.

From the out-of-control processes of motorization exceeding of admission possibility of street-travelling network on 20-30 reduces the actual carrying capacity of crossing in 2-3 times, up to the complete stop of a transport stream. For example, a middle rate of movement of transport on the highways of Russia is 40-60 km/h and 80-100 km/h abroad. Consequently, in Russia loads are moved for twenty-four hours to 250-300 kilometers against a 500-700 kilometer abroad. The decline of rate of movement, in turn, conduces to the increase on a 20-30 prime price of transportations, to the height of transport constituent in the last bid of products and services, that reaches in Russia to the 15-20 (in USA and Europe this index does not exceed 7-10%).

Basic backlogs of perfection of a transport process are in rational organization on the basis of engineering logistic of co-operation of all participants of transport-distributive chain of motion of load, in the concordance of their interests and search of взаимоприемлемых and mutually beneficial decisions. The logistic centers, created at the level of territories, municipal educations, metropolises, integrated branch complexes, can promote the decision of task of perfection of processes.

CT 2

It appears that a basic task of perfection of work of motor-car transport at moving of loads is reduction of general expenses in the process of satisfaction of transport necessities at maintenance of the accepted (or concerted) parameters of quality of service and quality of the state of environment.

In a region, under act of dynamically changing terms of work of transport vehicles, both in time and in space, there are processes of failure to observe of initial plan of a transport service, and insolvency of administrative influences shows up. All of it results in situations, when the existent models of vehicular process can not be effectively used for the operative management of decision of tasks for acceptable one time.

The ramified of street road net gives an wide opportunities for varying transport operations by optimization of routes of motion of transport vehicles between the points of service, local correction of these routes, by the redistribution of transport vehicles on the routes of delivery, minimization of expenses on providing of operative technical help or replacement of defective transport vehicles.

The really folded situation stipulated a management necessity transport streams on crossing of motorways with by the use of the traffic-light adjusting. The mathematical model of intellectual control system (ИСУ) is offered by transport streams of city on the basis of traffic-


light objects for reduction of expenses of resources of participants of process of delivery of loads. From position of approach of the systems a management process is examined by freight motor-car transportations under act of dynamically changing travelling, surrounding and external environments and terms of service.

In a market competition economy attention increases to the every family losses. The combined cost and time of delivery of load determine to a great extent, whether it will be maybe that or another economic co-operation. Id est a main value is presented by an integral function is delivery from a door to the door. But a function requires creation I correspond cabbage soup of structure that would provide her implementation.

Already clear, that the necessity of creation came to a head for the cities of ICS, that must provide the arranged technical co-operation of all participants of transporting and preparation of loads. Thus the functions of the system it is been: complete and timely reception of requests on transportation; calculation of the required park of rolling stock; serve of requests on providing of loading of empty cars; controller's support of transportation; conduct of the permanent monitoring of traffics of goods and determination of prospects of origin and redemption of traffics of goods; operative co-operating with the contiguous types of transport and a load is formed by enterprises.

Technology of work of regional ICS must be base on single through technological process of work of street-travelling network of city, envisaging the concerted admission of traffics of goods and rolling stock (carriages, cars) providing their on carriage.

CT 2 Operative management it is impossible a vehicular process to examine in tearing away

from a management by exploitation of roads and organization of travelling motion. The single technological process of work of street-travelling network of city must be based on satisfaction of economic interests of participants of vehicular process. The basic obligations of parties participating in transportation must be certain in him, sufficient in an order to serve as the regulator of their relations, and also to bear responsibility for their non-fulfillment.

Management in cities it maybe transport streams of street-travelling network to carry out due to the change of the modes of operations of traffic-light objects on the managed crossing, guided sign-boards. One of above all parameters of a transport stream on the basis of that made decision about the redistribution of green time between the groups of motion there is a delay of transport vehicles.

One of tasks of ICS is minimization of congestions on crossing of street-travelling network and increase of speed of delivery of loads. Existent model of control system on the basis of traffic-light objects can not transport streams help management centers to decide a main task travelling motion, and the considered variants of CAS of management by travelling motion by virtue of locality of application and by the difficulties, conditioned by absence of reliable methods of prognostication of distribution of transport streams at the different variants of project decisions a


System of logistic management freight motion on the street-travelling network of city is by the integrated system. The necessary volume of informative arrays part of that is local with the small period of storage is concentrated in her. Mathematical raising of task of optimization of advancement of transport streams on the basis of ICS formulated:

opt)V,O,K,R,A(Р1 → (1)

max)V,O,K,R,A(Р 2 → (2)

Where Р1- system of indexes of optimization of a transport activity(time, cost, normative troop landing in the environment of and other);

Р2 - a profit of logistic center of management for optimization of transport activity;

A - parameters of motor transport;

R - parameters of shifting complex transport of knot;

K - limitations of objects of infrastructure;

V - technological possibilities and limitations;

O - influence surrounding and external environments on transport processes.

A planning task is a multicriterion, depending on the large number of factors, and for every street-travelling network of region must decide separately coming from local terms and specific of them works. Thus planning must be not discrete, and continuous with providing of permanent depth of prognosis (for a 10-15 twenty-four hours forward and more), id est planning of traffics of goods from a moment and places of their origin and formations.

CT 2

As indexes must be worked out and accepted interdepartmental universal, balanced indexes, characterizing complex efficiency of cooperation (synchronizations) of proprietors of load, dispatch, and other participants of process of товародвижения in providing of transporting and processing of loads in a knot and on the way their following from the places of origin. As main indexes of choice most economically of effective variant of freight motion it is recommended to use time, cost of delivery and normative troop landing in surrounding environment.

minti

i →∑ (3)

∑ →i

i minS (4)

Where ti - time of transportation, clock;

Si - freightage, rubles;

Yi - normative extras in OS, conditional tons/year.

Three criteria create many criteria and needed methods of her overcoming. It is here possible to recommend two approaches.

First. Linear displacing. Three criteria are taken to one by means of coefficients of coercion.


Second. Hard priority. If one of criteria far excels by value the second, it is possible to examine them in turn. For example, the term of delivery costs on the first place. Then at first variants get out with minimum time of delivery, and among those, that is satisfied to the requirement of time, gets out with the least cost.

At a management travelling motion the structure of management object is known for and unchanging, and behavior of him depends on the row of unknown parameters. This task decides in the class of the self-tuning systems in that the structure of regulator is set and it is required to define the algorithm of tuning (algorithm of adaptation).

The task of synthesis of adaptive control system is considered. Let on the object of management (ОM) measureable indignations influence (questioner influences) of х(t), not measureable indignations of Q(t), surrounding and external environments of Fвi and Fпi and managing influences of U(t), influence on the object of management, fig 1.

CT 2

Fig 1. Flow diagram of adaptive control system by transport streams

Adjusting contour

Control system

Contour adaptations

External Surrounding environment environment FВi FПi

Filter

evaluations

measuring device

Managing system система

Algorithm adaptations

Ноб

Transport streams, ОM

Model

measuring devic

х(t)

Model of object

PS

Regulator

d

U(t)

Traffic-lights


A communication network includes a tie line, on that entrance information and reverse communication channel on that to the sensor-based system state information is passed management object are passed. Information about the guided object, external and surrounding environments perceived by the sensor-based system, processed in accordance with one or another aim of management and as managing influences passed through traffic-lights on the object of management (transport streams). A device for realization of purposeful influences is the sensor-based system, a management object and communication network form the system with a management (SM).

The supervisions a weekend is accessible to variables of object of management of Y(t). We suppose thus, that behavior of object depends on the row of unknown parameters totality of that is designated through D(t). In the process of management it is necessary to obtain freight motion of a transport stream travelling motion in the set mode, that will give: possibility to shorten time of passage on a network; to decrease the brought mass over of extrass of harmful substances in an atmosphere; to shorten transport delays and charges on moving of loads (passengers) and provide sufficient strength of travelling motion security. Because adaptive control system differ from the traditional (not adaptive) systems the presence of contour of adaptation, then for formalization of task of synthesis of algorithm of adaptation we use a concept "The Influenced object" (Ноб), that plugs in itself all unchanging part of the control system (СS).

If primary sensors (PS) do not provide measuring of necessary for a management parameters, such as speed, intensity, closeness of a transport stream, apply the filter of Кolmana, dynamic prognostication or methods of analysis of temporal rows.

CT 2

The algorithm of adaptive control has a three-level structure. The algorithm of the first level (adjusting algorithm or algorithm of base-level) depends on the vector of parameters of D (parameters of regulator), at each he must provide gaining end of management. Dd∈

An algorithm the second level changes(influences) the vector of D such by character, to provide gaining end of management at unknown Dd∈ .

On the basis of foregoing by us the three-level algorithm of adaptive control is offered by transport streams on the street-travelling network of cities. The algorithm of adjusting of a transport stream is modification of the coordinated management with introduction of new managing parameter - balance of time of resolvent and forbidding phase of the traffic-light adjusting ( ) and his rationed value. Management aim - the achievement of the set mode of motion of a transport stream (smoothing of rate of movement) on the highways of city is arrived at by the concordance of parameters of the traffic-light adjusting on a network.

10 ÷

The system described higher is a management task in the conditions of the vagueness related to , fig 1. We will consider the parameters of adaptive control system a transport stream of city. Entrance and output variables and managing influence are described as follows:

Dd∈


)B,V,Ц(F)t(U ,с3= (5)

)W,N,l,П(F)t(x cpр1= (6)

)Y,Ц,K,V,И(F)t(u сc2= (7)

Where Пр- carrying capacity of street-travelling network, cars/hour;

l - length of driving, m;

Nср - presence of technical equipments of adjusting, 1 or 0;

W - influence external and surrounding environments;

И - intensity of a transport stream, cars/hour;

V - speed of a transport stream, kilometer/an o'clock;

Кс - composition of a transport stream;

Цс - a cycle of the traffic-light adjusting, (seconds);

Y - ecological economic damage;

B - normative troop landing in an environment;

F1, F2, F3 - unknown functions.

In general case a management aim is set as having a special purpose inequality CT 2

0ТatТТ ллв >≤ (8)

Where Тв - time of passage on a street-travelling network, hour;

Тл - time of passage on a network at a local management.

As a regulator of adaptive control system a transport stream such technical equipments of adjusting as traffic-light devices and guided sign-boards are used. The vector of unknown parameters of d consists of traffic-light cycle indexes that is included in mathematical description of management object and correspond to the optimal tuning of the traffic-light adjusting:

)d,d,d,d,d(d 54321= (9)

Where d1 - duration of basic phase of adjusting (seconds) cycle;

d2 - duration of prohibitive phase of adjusting (seconds) cycle;

d3 - duration of intermediate times of adjusting (seconds) cycle;

d4 - balance of time of resolvent and forbidding phase of traffic-light cycle;

d5 - a change of phases of the traffic-light adjusting of "relatively command" traffic-light.


The algorithm of adaptive control is offered by transport streams on the street-travelling network of cities with the use of the traffic-light adjusting, taking into account an ecological factor. The algorithm of adjusting of a transport stream is modification of the coordinated management with introduction of new managing parameters - balance of time of resolvent and forbidding phase of the traffic-light adjusting (0÷1) and his rationed value and account of extras of harmful substances in an environment. Management aim - the achievement of the set mode of motion a transport stream smoothing of rate of movement) on the highways of city and normative troop landing in an environment is arrived at by the concordance of parameters of the traffic-light adjusting on a network.

Because of territorial disconnect, and also distinction of parameters of functioning, for effective work of freight transport there must be a single center of management, in that in good time and operatively must act information about performance of every subject of a transport stream indicators. Adjustment of strategy of development of a transport stream, and also, in a short-term prospect, change of managing influences and redistribution of resources, must be executed taking into account the results of analysis of acting operative information. At the same time, as co-operation between subsystems comes true by means of informative and material streams, her it is also possible to examine as intellectual control system, id est adaptive system ticker-coil, where, where informative streams act part managers and correct of description material streams.

Guidance of enterprises tests a requirement in reliable information, as on it management quality depends by an enterprise and efficiency of planning of his activity in the conditions of hard competitive activity. For informative support of every stage of management application of ICS providing the choice of effective administrative decisions and operation ability of reaction on the market state of affairs and changes external and surrounding environments is needed. Approach of the systems to the choice of rational management must be used on all stages. On every stage the characteristic are used for the decision of similar tasks methods and models.

CT 2

The offered algorithms of management give an opportunity to shorten time of passage on a network freight motion, to decrease the extras of harmful substances in an atmosphere, transport delays and charges on moving of loads and passengers.

References

[1]. Korchagin V.A. Innovative ecological econom.-Lipetsk / LGTU, 2010. – 200p.

[2]. Korchagin V.A. Model of search of the effective functioning of motor transport natural economic frame//of society the World transport and technological machines. - 2011. - 2. - With. 25-31♦


SUSTAINABLE DEVELOPING ROAD TRANSPORT SYSTEM FOR HANOI

TRAN TUAN HIEP TRAN VU TUAN PHAN Civil Engineering Faculty, UTC

Summary: This presents characteristic Road transport System of Hanoi, and propose the objective, principles and solutions for sustainable developing.

Key word: Hanoi, Road transport, sustainable developing.

I. OVERVIEW

In general, most of the urban in the world has been self-developed, derived from the factors that support activities, manufacturing and transportation. Those are elements of geography, soil, location and population, etc. For example:

- The confluences (supporting water way exchanges). - The rivers, mountains and forests (supporting raw materials, fuel and trading). - Delta area (supporting cultivation, transportation, fertilizing soil, construction, expansion, etc.). - The location adjacent with the territories... Such a place with such favorable natural conditions has been chosen to be where to form

city’s center. Over time, with the development of the economy-society, commercial exchange develops more while pushing the development of transport system forming as centripetal (Radial axle roads and urban ring roads).

Most of such urban has the central area with deep antique impression; urban density is very high, crowded in the hub, stretching from inside to outside. Urban develops by the expanding and invading type.

There is also young urban (with its road network) which develops in the form of fish bones, chessboard, chain or mixture.

II. DEVELOPING ROAD TRANSPORT SYSTEM - OBJECTIVE & PRINCIPLES

Objectives Road Transport System is a special system which consists of both entities structures and

regimes to operate it to serve the transport through, safely, smoothly, comfortably, economically and aesthetically.

“Through” means: transport must be normally continuous during space: over service areas; during time: over seasons without influence of climate or natural condition. To meet the people demand of door-to-door and just in time

“Safely, smoothly, comfortably” show the service quality.


“Economically” expresses the effectiveness of all system which achieved by infrastructure, transport means, legislation, and policies to manage and operate the system.

“Aesthetically” in accompany with safely, smoothly, comfortably response the requirement, being friendly with environment and people.

Principles

• Urban transport master plan has to be based on and originally generated from general planning for economy-society city development, urban space development and land use.

• Planning has to be based on the long term strategic vision, at least for 50-100 years later.

• Aiming all the effort to develop economy-society, for human, for the development of human, for the harmony between people and people, people and nature, society; this means for a healthy economy-socio-cultural and highly developed human environment.

• Developing road transport based on smart growth, reasonable balance all modals of transport, priority for ITS and environment friendly technology.

• Transport system along with urban infrastructure has to go a step ahead in order to lay a foundation and motivation for urban sustainable development.

• The heritage, traditional value must be inherited, reserved and promoted, we should synchronously acquire and study the advance achievements and experiment of world’s urban in the globalizing cooperation, affiliation and development.

• The urban of developed countries has to base on the characteristics of the region; inherit, preserve, promote and study to find out their solutions in order to develop quickly and sustainable with the criterion: “Tradition, Nation and Modern”.

III. TREND TO DEVELOP ROAD TRANSPORT SYSTEM

Born earliest and expanded largest, Road Transport has been being increasingly developed all over the world.

In developed countries, now they tend to find out advanced breakthrough technology for better Road Transport in future:

- To create “greener” road transport. The European Road Transport Advisory Council (ETRAC) has proposed for a directive on promotion of clean and energy efficient vehicles (2007)

- To encourage modal shift and decongesting transport corridors. - To ensure sustainable urban mobility - To improve safety and security - To strengthen competitiveness These trend lead to the concept of developing and integrated transport system, Road

transport and health, request to develop the ITS. ULTra (urban light transport) in European seems a breakthrough technology of road

INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04 12

transport. This was applied at Heathrow airport as pilot project and shown many benefits: - Immediate service: passengers rarely need to wait for a vehicle. Since the empty vehicle

management system ensures that one will already be at the station. Simulations demonstrate that average waiting times, even in peak periods, average around 10 seconds.

- Nonstop travel: due to off-line stations, the journey is nonstop from start to destination, anywhere on the network.

- No need to plan trips, considers schedules, or transfers between vehicles. - Your own private cab you only share with your friends and family. - Faster than other urban transport, typically by a factor of two or three. Although

maximum speeds are modest (25 mph), non-stop service ensures short trip times. - Travel is reliable, predictable and congestion free affording passengers greater certainty

in their journeys. - Travel is safe: ULTra’s target is safety levels at least as good as trains, approximately 10

times higher than automotive safety. Also segregation implies less conflict with non-users. - Accessibility: The system is available to all, including the young, the old, and those with

disability. In addition to user benefits, ULTra provides sustainable urban transport with major benefits

to non-users and to the community as a whole. - ULTra is energy efficient: Light, small, efficient vehicles traveling non-stop and only on

demand result in significant energy savings. ULTra saves 2/3 rds of automotive energy requirements, and is substantially more energy efficient than conventional public transport.

- ULTra meets Kyoto sustainability targets; providing the required 60% reduction in carbon emissions over the car now, rather than by the 2050 target date of the Kyoto agreement – 35 years ahead.

- ULTra is exceptionally quiet: measurements on the prototype vehicle running at 6m/s give 35 dBA at 10m, around 30 dB less than cars.

- Light weight vehicles permit ultra-light infrastructure: Automated control allows high utilization. Small vehicles and guide ways imply less land take and less visual instruction.

- ULTra reduces congestion: Studies indicate significant modal shifts away from the car, freeing up both road capacity and parking space.

- Installation flexibility: Small scale infrastructure may be readily integrated into buildings. ULTra provides new ways to reclaim areas of the city now given over to the automobile.

- Rapid installation minimizes cost and disruption.

IV. SOLUTIONS FOR HANOI URBAN ROAD TRANSPORT SYSTEM TOWARD THE SUSTAINABLE URBAN DEVELOPMENT

4.1. Characteristics of Hanoi urban road network Hanoi has the center lays on the confluences of Duong River --- Hong River on the Delta

which has flat, large and rich land. With the spontaneous development of urban for more than


1000 years until now, this place forms a crowded urban with the population of no more than 10km2 (road transport system is primary, waterway is less developed, railway and airway do not serve external transport relationship).

• Road network - Radial axle and ring road system merely developed (have just formed 3-4 lanes in the

inner city but hasn’t constructed and rehabilitated completely, hasn’t gotten suburb’s ring roads). Therefore, external transports, national transport from different directions have to pass through the center of city making it pressed, overloaded and congestion for the inner city.

- The streets mostly are two ways, the carriage way is narrow. - The sidewalk is narrow, patchy and out of standardization. - The inner city intersection system is at grade, much and closed to each other (200-400m). - Residents’ houses system inserts heavily along the inner city’s street, is both living and

settling place, manufacturing, trading, etc. since long time ago; thus, it is very hard to clear away, compensate and resettle in order to widen and improve the road (also build and improve inner city’s public works).

- The urban system and the suburb’s road network are merely developed. - The system of Red river bridges is not enough capacity. There are only 3 bridges: Thang

Long, Chuong Duong and Long Bien old bridge. - The urban transport system is simple, lack of synchronous development of railway and

waterway.

• The parking system - The parking system is poor. The shops system crowded along the streets, small markets

and schools, etc. which create the problem that the traffic vehicles (cars, motorcycles…) occupy pavement and sidewalk for parking so that the narrow streets become narrower.

• The transport means - The transport means are mainly cars, motorcycles and bicycles. Public transportation is

mainly bus, is given especial encouragement in 2000-2005, however just serve about 20% of the travel demand but cannot develop because the road system is not satisfied yet.

• The motorcycle boom and mix traffic flow

- There may be no such an urban in the world that has a many motorcycles as in Hanoi; motorcycles serve more than 60% the travel demand of urban residents.

Until November, 2005, Hanoi has nearly 163,000 cars (with the increasing rate of 12-14% per year), more than 1,5 million motorbikes (with the increasing rate of 14-16%) and more than 1 million bicycles (has the trend to a slight decrease).

Bicycles, motorcycles and cars put the mix traffic flow on the road is a outstanding characteristic of Hanoi urban transport system, is currently a hard problem to the managers and controller of urban transport.


• The transport participants - Almost Hanoi transport’s participants are not only the passengers but also the transport’s

controller (which means also the drivers). Coming from the habit of agricultural economy, small manufacturing; the knowledge, the awareness for abiding by traffic regulations is not quite good. There are even people who just care about when can go but need not to know when can be go.

• The institution system, urban transport control and management: Not really explicit and completed.

Fig. 8.4. Hanoi Road network

Fig. 8.5. Occupy carriageway, sidewalk for parking Fig. 8.6. Mix traffic, and traffic jam

4.2. Solutions for Hanoi

• Hanoi chooses its manner to develop from outside to inside - Basic argument As above mentioned, like many cities which has a long spontaneously developing process

has created Hanoi, crowded and overloaded central area. The difficulty of the clearance work and relocation for improving and rehabilitating the interior area, the conflict between reservation and development is the biggest impediment of the sustainable urban development.


Hanoi does not develop evenly; the suburb is currently a poor rural condition. It is a favorable condition for constructing urban infrastructure synchronously and modernly.

The pressure of the internal area creates the inhabitant’s requirement for expanding to the borders, which increases the trend of purchasing lands to build houses in the suburban area. This creates sprawl construction in the suburb which is over controlled.

Many big cities in over the world have developed from inside to outside. Hanoi developed slowly, and is still on its long way after other cities and urban, if goes on the same way, Hanoi can never catch them. The problem here is to find its own way, which is to develop urban from exterior into interior area.

- Concrete solutions: First, to limit temporarily the development of the internal area. Concentrating the sources on suburb’s development, building urban infrastructure to be

modern and synchronize. Specifically develop the radial road axle system and outer ring roads with 15; 20; 25; 30 km far from the center as express way; next are to develop branch roads and local roads into standard as well. Based on that, establishing the urban functional area can be established to be modern as its planning.

Pulling and obligating some organs, universities, schools and administration offices from internal area to exterior area.

Then reserving, improving and regenerating the interior area into a political, culture, service, commercial, tourism, antique, modern and peaceful area.

- Hanoi axle road network proposal.

Fig. 8.7. Hanoi axle network proposal


• Hanoi certainly has to develop public transport system including buses and urban railway system (metro). Within 15 year coming, Hanoi must to build at least 20-30 km of metro

• Hanoi certainly has to eliminate motorcycles Hanoi’s actual development has raised the problem of motorcycle explosion, this traffic

type develops so fast that makes the urban management over controlled, is also a reason for traffic jams and traffic accidents.

Based on that fact, there have been many measures like increasing consuming tax, prohibiting people from registering motorcycles in the inner districts, improving and investing equipment; executing administrative and educational solutions, etc.; however, the numbers of motorcycles running in the city have still been increased, the solutions are not effective yet.

- It is the time to answer definitively; (1) Can Hanoi eliminate motorcycles?, (2) When will Hanoi eliminate motorcycles?

- Answer for question number (1): Show the decisive determination of the top leader to attract and focus the sympathy power of the society and the people.

- Answer for question number (2): Determine the exact time for researchers, urban managers studying to sketch out a reasonable process and comprehensive, synchronous and feasible solutions for eliminate motorbike. Pay a special attention for the developing public transport and dominate steps by steps private vehicles; so that urban residents have time to be mentally prepared and choose the suitable solution for them and their own family.

We think that: in 2020, Hanoi has to say “No” to motorcycles. 15 years are enough to step by step to build and develop public transport, step by step limit and then eliminate motorcycles, simultaneously execute synchronous solutions for the economy and the society. If the progress is too fast, it will be unfeasible; if it is too slow, motorcycles will develop until the level it cannot be eliminated, causes bad affection to the urban development unable to be solved. Thus, the urban planners and managers will be unsuccessful.

• To invest and construct parking system to be enough and reasonable for the city. • Developing a special transport system, “smooth transport” by using traffic means

unpolluted and by establishing walking streets in the commercial and tourist area, old streets, traditional and cultural handicraft villages.

• To strengthen road transport regulations system and related policy in order to make more effective urban transport management as well as providing, implementing administrative and educational solutions to improve knowledge and awareness to obeying traffic regulations of urban residents.

Besides road transport and urban railway, Hanoi also has a system of diversified rivers and lakes system; thus, waterway transport serving tourism, as air transport for external relations and international integrated transport are significant factors in developing our capital to be modern and sustainable.

References [1]. Martin V. Lowson, APRT Solusion for Europea Cities, Infrastructure Feb. 2007 [2]. ULTra Website, Advanced Transport System Ltd♦


INFLUENCE OF AERODYNAMIC CONFIGURATION OF A STREAMLINE BOX GIRDER ON BRIDGE FLUTTER AND

VORTEX-INDUCED VIBRATION

QI WANG HAILI LIAO MINGSHUI LI CUNMING MA

Research Center for Wind Engineering, Southwest Jiaotong University, Chengdu 610031, China

Summary: Streamline box girders are widely applied in the design and construction of long-span bridges all over the world. In order to study the influence of modifications of aerodynamic configuration and accessory components on flutter and vortex-induced vibration (VIV), more than 60 cases were tested through a 1 scale section model. The test results 50׃indicates that the aerodynamic configuration and accessory components of streamline box girders can significantly affect the wind-induced vibration of bridge, which is in good agreement with the experience of past researchers. From the tests carried out, it is observed that if the horizontal angle of the inclined web of the streamline box girder is below 16°, the critical flutter wind speed of bridge will increase remarkably, and the VIV will diminish. The test results also show that the 15° inclined web can restrain the formation of vortex near the tail, and consequently improve the performance of aerodynamic stability of long-span bridges. Finally, a new streamline box girder with 15° inclined web was presented and strongly recommended in the aerodynamic configuration design of long-span bridges.

Key words: streamline box girder; aerodynamic configuration; wind tunnel test; flutter; vortex-induced vibration.

I. INTRODUCTION

Since its successful application to the Severn Bridge of Great Britain, the streamline box girder has been widely used in the design and construction of long-span bridges all over the world. Its aerodynamic configuration and accessory components have great impacts on bridge wind-induced vibration, especially the flutter and vortex-induced vibration (VIV). Larsen [1] conducted a series of wind tunnel tests on the girder of the Great Belt Bridge, and discussed the influence of railing, guide wing, and rostrum on the flutter. He found that the railings of low porosity weakened the aerodynamic stability, the blunt rostra decreased the flutter wind speed, and the guide wing strengthened the aerodynamic stability. Similar results were obtained by Miyata [2] and Bruno [3]. Wilde et al. [4] designed an active deck flaps control system to strengthen the aerodynamic stability of long-span bridges, but this system could increase the construction and maintenance cost of bridges, and its feasibility and efficiency in real-world applications still need to be validated. Yang and Ge [5] studied the effects of central stabilizer on aerodynamic control of long-span bridges. The results indicated that the stabilizer did good to aerodynamic stability, and its fixed position must be confirmed by wind tunnel tests.


However, the stabilizer did not fit for a traditional box girder. Through the wind tunnel tests and experiments [1]-[5], it was found that the flutter of long-span bridges could be avoided by optimizing the aerodynamic configuration and deck details. Compared with the studies on flutter of the streamline box girders of long-span bridges, however, the studies on VIV are not often seen. Larsen et al. [6] discussed the guide vane of box girder for suppression of the VIV of the Great Belt Bridge. He provided a general design of the guide vane for the twin box girder of Stonecutter Bridge in Hongkong [7], but this design would raise the construction cost. Diana et al. [8] investigated the vortex-shedding phenomena of the multi-box deck shape of Messina Strait Bridge and discovered the remarkable VIV phenomena in the bridge, but he did not provide the effective measures to suppress the VIV of the triplet girder. Ge et al. [9] carried out wind tunnel tests and field tests for the VIV of Xihoumen Bridge with a twin box girder, and found that the windbreak could control VIV effectively but would increase the drag force of girder and weaken the aerodynamic stability of the bridge. The aforementioned work indicates that there are lack of effective and practical measures to suppress VIV and maintain the aerodynamic stability of long-span bridges.

In order to clarify the influence of accessory components and aerodynamic configuration on the aerodynamic stability and VIV of long-span bridges, this paper deals with wind tunnel tests through a 1 scale section model of streamline box girder. More than 60 cases of deck 50׃configuration were tested, such as the bridge railings with different porosities in the sideway, the position of the rail of bridge inspection car, the obliquity and width of guide wing, the edge configuration of the section (rostrum), and the inclination of the inclined web. Based on the tests, a new design for aerodynamic configuration of bridge girder is put forward, which could ensure the superior aerodynamic stability and less VIV of long-span bridges.

II. WIND TUNNEL TEST MODEL

The design girder cross-section of the Nanjing 4th bridge, which crosses the Yangtze River and is located in Jiangsu Province of China, with the main span of 1 418 m, was selected to carry out wind tunnel tests. The bridge deck is originally designed a trapezoidal steel box girder with overall width of 37.7 m and a height of 3.4 m, and the inclination of inclined web is 22°, as shown in fig 1. For long-span bridges, the aerodynamic stability is a governing factor in the design. According to the wind statistic data and the Wind-Resistant Design Specification for Highway Bridges [10], the flutter checking wind speed of the bridge is calculated as up to 60.8 m/s. However, the critical flutter wind speed of the original girder is found by intensive wind tunnel testing of the section model only 45 m/s. Apparently it falls short of the required flutter checking wind speed. Hence, in the aerodynamic design of this bridge, it is necessary to adopt some countermeasures to ensure the critical flutter wind speed meets the requirements of wind resistance design. And on the other hand, we attempt to find the necessary measures to suppress the VIV through the tests.

Fig 1. Outline of the streamline box girder in the tests


The wind tunnel tests were conducted in the second test section of XNJD-1 wind tunnel. The model is 2.1 m long, and its width and height vary with the change of aerodynamic shape. The mass of the model is 23.52 kg, and the mass moment of inertia is 1.407 kg·m2. The vertical and torsional frequencies of the test system are 1.76 Hz and 4.03 Hz, respectively, and the wind speed ratio between model and real bridge is 3.28. The damping ratios of the test system in vertical and torsional directions are 0.45 and 0.48, respectively. The aeroelastic section model in the wind tunnel tests is shown in fig 2.

Fig 2. Section model in the wind tunnel tests

III. INFLUENCE OF BRIDGE RAILING

The railings of two porosities (90% and 60%) at three different locations on deck were tested under three attack angles (−3°, 0° and 3°). The three locations are as follows: one is on the edge (the original design), the other two are 5 mm (prototype 250 mm) and 10 mm (prototype 500 mm) away from the sideway edge. The test results are given in tab 1.

Tab 1. Critical flutter wind speeds for different railings (m/s) Attack angle Porosity Position

−3° 0° +3° 90% Initial position >74.6 73.0 61.3

60% Initial position >75.3 57.6 44.8

60% 250 mm inside >74.8 58.6 42.4

60% 500 mm inside >76.2 57.9 44.0

We find that the railings with high porosity strengthen the aerodynamic stability. Similar findings were also reported in refs. [1]-[3]. The railings of low porosity mean the deck configuration being in the form of an H-shape, which may lead to flow separation. Then a rhythmic vortex shedding will lead to the generation of separation bubbles above the deck. The vortex creation and drift process will not only dramatically weaken the stability of girder, but also increase the amplitude of VIV [11]-[12]. The test results in tab 1 indicate that the railings of low porosity weaken the aerodynamic performance of bridge girder. However, the railings of high porosity are harmful to the safety and durability of long-span bridges. Therefore, in the design of railings, both the safety of bridge and the aerodynamic performance of the bridge girder should be ensured. Tab1 also shows that the critical wind speed of the girder is not sensitive to the railing position change.


IV. INFLUENCE OF THE RAIL OF BRIDGE INSPECTION CAR

The aerodynamic stability of the girder may be sensitive to the distance between the rail top and the girder bottom, called the rail gap, and also sensitive to the fixed position of rail on the girder. Tests on three gaps were carried out to investigate the potential influence. The gap was increased by 1 mm and 3 mm (prototype 5 cm and 15 cm) in the tests (the porosity of railings is 90%). Flutter critical speeds of these sections were measured, and are summarized in tab 2. It is found that increasing gap is beneficial to the critical wind speed at 0° and −3° attack angle, but harmful to the aerodynamic stability of the girder at +3° attack angle. It can be seen from tab 2 that the critical flutter wind speed decreases with an increase in the rail gap. Thus, increasing the rail gap is not a good choice to strengthen the aerodynamic stability of the girder.

Tab 2. Critical flutter wind speed under different gap (m/s) Attack angle Case Position

−3° 0° +3°

Initial design

5 mm from the bottom >79.0 73 61.3

Increment of 1 mm

6 mm from the bottom >81 >80 59.6

Increment of 3 mm

8 mm from the bottom >84 >82.5 58.1

The sketch maps for three different fixed positions of rails are shown in fig 3, and the VIV test results of the three cases under +3° attack angle are shown in fig 4. If the rail was close to the corner of the girder bottom (see fig 3-(a)), which is the separation point of incoming flow, the amplitude of VIV would become too large to exceed the code limit. If the rail was far from the corner of the girder bottom (see fig 3-(b)), the amplitude of VIV would become small, or tend to zero. If the guide vane was fixed at both sides of the rail in the original design, shown in fig 3-(c), the VIV amplitude could decrease significantly. The position of rail has a large effect on the VIV.

800

4 800

(a) Original design

800

(b) The adjusted position of rail

(c) Original design with guide vane Fig 3. Positions of the rail of bridge inspection car (unit: mm)


2 4 6 80

50

100

150

200

10

Original position Adjusted position Original position

with guide vane

Am

plitu

de o

f V

IV (m

m)

Wind speed (m/s) Fig 4. VIV amplitudes for different rail positions

V. INFLUENCE OF GUIDE WING

The guide wing on the edge of sideway, shown in fig 5, can smoothen the airflow passing through the section, which may strengthen the aerodynamic stability [1]-[3]. A total of nine different types of guide wings were employed in the intensive wind tunnel tests, with different widths and obliquities. The railing porosity was 60% for the section model in the tests. Flutter wind speeds were obtained in the tests for the section model under attack angles of +3° and 0°, and the results are shown in tab 3.

Tab 3. Critical flutter wind speed with guide wings (m/s)

Guide wing Attack angle Width (cm) Obliquity (°) 0° 3°

50 +15 52.2 44.9 50 0 51.47 41.98

50 −42 54.75 42.34

100 +15 54.96 52.05 100 0 52.78 50.23

100 −42 58.77 45.63

125 +15 60.23 63.88 125 0 56.94 55.48

125 −42 62.78 49.64

It is observed from tab 3 that, with an increase in the width of the guide wing with a positive obliquity, the aerodynamic stability is strengthened even though the railings have a low porosity. However, the guide wing will increase the complexity of the structure and its construction, particularly in the location of rostra, and the maintenance cost will increase accordingly. Therefore, a guide wing fixed at girder sides is not recommended in the design unless there is no alternative.


ObliquityWidth

Guide Girder

Fig 5. Outline of guide wing

(a) (b)

VI. INFLUENCE OF ROSTRUM

The critical wind speed is sensitive to the shape of the rostrum [1]-[3], [11]. Therefore, the rostrum with different widths and angles was taken into account in the tests. The angle of the rostrum varied from 57° to 25°; correspondingly, the width varied from 1.9 m to 3.3 m. A total of 21 model cases were tested. The results are shown in tab 4.

Tab 4. Critical flutter wind speeds with different rostra (m/s)

Rostrum Attack angle

Angle (°) Width (m) +3° −3° −3°

57 1.9 51.5 >69.6 >70.6

47 2.1 53.7 >70.7 >70.3

41 2.3 54.6 >71.2 >70.5

37 2.5 56.7 >70.8 >71.2

33 2.7 58.6 >71.5 >75.2

30 3.0 63.4 >72.5 >70.9

25 3.3 59.3 >71.3 >72.8

It is noted that with an increase in the rostrum width and a decrease in the rostrum angle, the critical flutter wind speed increases. However, the critical flutter wind speed declines when the width of rostrum exceeds 3 m, which may weaken the stability, and the reason will be discussed in the later section of this paper.

Similar conclusions were obtained from the VIV tests. The amplitude of the VIV decreases with a decrease in the angle of the rostrum (see fig 6). Thus, it can be inferred that the blunter the rostrum, the larger the VIV amplitude. The reason is that being the first separation point of the incoming flow, when the rostrum becomes blunter, the flow is easier to separate, and the vortex will form and drift over the girder, which can intensify the VIV amplitude. According to this result, a blunter rostrum, or the similar aerodynamic components which make the rostrum blunter, such as the ordinary sidewalk board, is not recommended in the design.\


2 4 6

0

50

100

150

200

250

8

57° rostrum 41° rostrum 30° rostrum

Am

plitu

de o

f V

IV (m

m)

Wind speed (m/s) Fig 6. VIV test results with different rostra

VII. INFLUENCE OF INCLINED WEB

A wide and acuminate rostrum is difficult to fabricate and fix, implying more cost in design and construction, although it can strengthen the aerodynamic stability of the girder. An alternate way is to decrease the steepness of sidewall slope to make the deck cross-section more streamlined. The critical flutter wind speed of the girder model with different slopes of the inclined web is shown in tab 5 (the width of rostrum was 2.4 m), and the value increases with the decrease of the slope. When the slope is decreased to 15°, the critical flutter wind speed rose to 67.1 m/s, directly improving the flutter instability by 10%. From tab 5, we can obtain an ideal section of girder for improving flutter instability: short rostrum, no guide wing, and railings of low porosity.

Tab 5. Critical flutter wind speed with different slopes (m/s)

Attack angle Slope of inclined web (°)

+3° 0° −3°

22 44.8 57.6 >75.3

20 56.7 >70.8 >71.2

18 60.4 >72.5 >70.9

15 67.1 >71.3 >73.5

In addition, we have a new finding in the VIV tests. When the slope of inclined web is decreased to 15°, without the railings and the rail of the bridge inspection car, there is no VIV phenomenon. The results of VIV tests in different cases are shown in fig 7, where the slope of inclined web is 22° for original design, 15° for the adjustments without railings and without rail. Compared with the original design, the bare girder section without aerodynamic accessory components does not experience VIV. There is a clear conclusion that the rail of the inspection car on the bottom of girder will intensify VIV. The same conclusion was reached by Larsen [14] in 2008, who first discovered and documented this phenomenon. Furthermore, he verified this finding by considering two other bridges.


2 4 6 80

50

100

150

200

10

Original design Without railing Without rail

Am

plitu

de o

f V

IV (m

m)

Wind speed (m/s) Fig 7. Comparison between different VIV test results

VIII. DISCUSSION ON AERODYNAMIC PERFORMANCE IMPROVEMENT

Based on the experience from vortex shedding tests, Larsen [13] found that the streamline box sections were similar to the airfoils at aerodynamic performance, and concluded that the flow along the bottom plate would stay mainly attached if the slope of inclined web was less than approximately 16°. He conducted a series of special tests and verified the above conclusion through the flow visualization technique [14]. He also gave an example about the design for a two span suspension bridge in Chile, whose inclined web angle was 14.8°, and no VIV was observed in the wind tunnel testing. On the contrary, explicit VIV was observed for the box girders of the Great Belt East Bridge and Osterøy Bridge, whose inclined web slope angles were 26.6° and 29.5°, respectively [14].

By analyzing Larsen’s test results and the test results of fig 7, we can draw the conclusion: vortex shedding excitation originated from rhythmic vortex formation under the downwind inclined web of the box section can be eliminated by choosing the slope of inclined web at about 15° and shielding rail (see the test case without rail in fig 7).

Similar to a box section in the vortex shedding vibration status, in the flutter critical status, two vortices with opposite directions were observed at the both sides of the nose-tail line of Great Belt East Bridge section through the PIV (particle image velocimetry) technique [15]. The two vortices could give the girder enough momentum to increase its VIV amplitude in a short time and finally made the girder instable. When the wind speed was low and below the flutter critical speed, the positive vortex below the nose-tail line was more powerful than the above negative one, and the aerodynamic force was just a static lift force; when the wind speed increased close to the flutter critical speed, the negative vortex above the nose-tail line was strengthened to a level as powerful as the positive one below, and the aerodynamic force fluctuated [15]. If the frequency and the phase of the fluctuated force induced by vortex shedding are close to the bridge’s modal frequency and phase, the flutter of girder will occur soon.

In the tests conducted in this paper, when the slope of the inclined web is decreased to 15°, there is a small dead wake region below the nose-tail line and the flow along the bottom plate


stays mainly attached to the web, which makes it more difficult for formation of a large vortex. When the wind speed increases and approaches to the original critical flutter wind speed (45 m/s), the opposite vortices can not become powerful, and can not give the girder powerful and efficient excitation. Fig. 8 shows the two vortices near the tail of the girder at this status. They are so asymmetrical that they can not produce the fluctuated aerodynamic force, and thus, the bridge remains stable. When the wind speed continues to increase, the equivalent opposite vortices will form at the similar position and the rhythmic excitation will return, which introduce aerodynamic instability to the girder once more. Fig. 9 reflects the status of the vortex moment at the wind speed approaching to the flutter critical value, and the equivalent vortices near the tail of the girder can produce the fluctuated aerodynamic force that makes the bridge instable.

Fig 8. Vortex moment at the low wind speed (below the critical value)

Fig 9. Vortex moment when the wind speed approaches to the critical value

Therefore, vortex shedding excitation originated from the equivalent opposite vortices near the tail of the box section can be restrained and delayed by choosing the inclined web angle at 15°. The test results in tab 5 can verify this conclusion. In aerodynamic stability design of long-span bridges, to increase the critical flutter wind speed, the same design of the bridge girder with the inclined web at 15° is recommended.

This explanation can also be extended to interpret the results shown in tab 4. When the slope of the inclined web is less than 16°, the increase of the width of the rostrum leads to the formation of a large vortex, which is a disadvantage to the girder stability and make the critical flutter wind speed decrease. Tab 6 shows the test results of the critical flutter wind speed with different width of rostra for inclined web of 15°. We can conclude that the short rostra are of benefit to the girder stability when the slope of inclined web is 15°.


Tab 6. Critical flutter wind speed (slope 15°) (m/s)

Rostra Attack angle

Width (m) Angle (°) +3° 0° −3°

2.4 40 67.1 >71.3 >73.5

2.6 36 62.0 >70.8 >72.7

2.8 33 61.1 >71.5 >72.4.

It should be noted that the explanation documented above is based on the Larsen’s research [14] on the vortex shedding vibration of streamlined box girder, and Zhang’s research [15] on the flutter critical status of Great Belt East Bridge by PIV technique. The explanation is obtained by a logical deduction about aerodynamic stability improvement. It needs to be verified by the further study of the wind tunnel tests using PIV technique and computational fluid dynamics method in the future.

IX. CONCLUSIONS

In this paper, more than 60 cases were tested through a 1׃ 50 scale section model to study the influence of aerodynamic configuration and accessory components on the flutter and vortex-induced vibration (VIV), such as the railing, the rail of inspection car, the rostrum, the guide wing, and the slope of inclined web. From the tests, we can obtain several important conclusions for the aerodynamic design of streamline box girder in long-span bridges. The girders with railings of high porosity have higher critical flutter wind speed than the ones with railings of low porosity. The wide and acutance rostrum can strengthen the aerodynamic stability of the girder, but the width of rostrum cannot exceed 3 m. The rail of the bridge inspection car fixed at the corner of bottom plate of box girder can intensify VIV. It is recommended that the rail should to be shielded by guide vanes.

If the inclined web with a 15° angle is employed in the section design of long-span bridges, the flutter performance can be enhanced or the critical flutter wind speed can be increased dramatically. The VIV of long-span bridges under the practical structural damping can be also suppressed by this same design.

Through the wind tunnel tests, we find that the final design section of the Nanjing 4th bridge, including the railings of 60% porosity, shielded inspection car rail, inclined web with a slope of 15° and 2.4 m wide rostrum, has a good aerodynamic stability, and the critical flutter wind speed is up to 67.1 m/s. Of course, there is no vortex shedding vibration observed in the 1 ׃ and 1 50׃ 20 section model tests. This section model offers a good demonstration for the aerodynamic configuration designs of other long-span bridges. The points of this paper have been


validated by the aerodynamic configuration design of Huangyi Bridge crossing the Yangtze River, which is located in Luzhou City, Sichuan Province.

References [1]. A. Larsen, Aerodynamic aspects of the final design of the 1 624 m suspension bridge across the Great Belt, Journal of Wind Engineering and Industrial Aerodynamics, 1993, 48(2-3): 261-285. [2]. T. Miyata, Historical view of long-span bridge aerodynamics, Journal of Wind Engineering and Industrial Aerodynamics, 2003, 91(12-15): 1393-1410. [3]. L. Bruno, G. Mancini, Importance of deck details in bridge aerodynamics, Structural Engineering International, 2002, 12(4): 289-294. [4]. K. Wilde, P. Omenzetter, Y. Fujino, Suppression of bridge flutter by active deck-flaps control system, Journal of Engineering Mechanics, 2001, 127(1): 80-89. [5]. Y.X. Yang, Y.J. Ge, Some practices on aerodynamic flutter control for long-span cable supported bridges, In: Proceedings of the 4th International Conference on Advances in Wind and Structures (AWAS’08), Jeju, Korea, May 28-30, 2008. [6]. A. Larsen, S. Esdahl, J.E. Andersen, et al., Storebaelt suspension bridge-vortex shedding excitation and mitigation by guide vanes, Journal of Wind Engineering and Industrial Aerodynamics, 2000, 88(2-3): 283-296. [7]. A. Larsen, M. Savage, A. Lafrenière, et al., Investigation of vortex response of a twin box bridge section at high and low Reynolds numbers, Journal of Wind Engineering and Industrial Aerodynamics, 2008, 96(6-7): 934-944. [8]. G. Diana, F. Resta, M. Belloli, et al., On the vortex shedding forcing on suspension bridge deck, Journal of Wind Engineering and Industrial Aerodynamics, 2006, 94(5): 341-363. [9]. Y.J. Ge, Y.X. Yang, F.C. Cao, VIV sectional model testing and field measurement of xihoumen suspension bridge with twin box girder, http://www.dist.unina.it/proc/2011/ICWE13/start/papers/502_8page_viv_sectional_model_testing_and_field_measurement_of_xihoumen_suspension_bridge_with_twin_box_girder.pdf, 2011-10-30. [10]. JTGT D60-01-2004, Wind-Resistant Design Specification for Highway Bridges. [11]. J.Z. Song, Z.X. Lin, J.Y. Xu, Research and appliance of aerodynamic measure s about wind resistance of bridges, Journal of Tongji University (Natural Science), 2002, 30(5): 618-621 (in Chinese). [12]. C.J. Liu, Z.S. Guo, L.D. Zhu, Influence of railing curbstone structure on flutter stability of box main girder, Bridge Construction, 2008(2): 20-22,44 (in Chinese). [13]. A. Larsen, Aerodynamic stability and vortex shedding excitation of suspension bridges, In: Proceedings of the Fourth International. Conference on Advances in Wind and Structures (AWAS’08), Jeju, Korea, 2008. [14]. A. Larsen, A. Wall, Shaping of bridge box girders to avoid vortex shedding response, http://www.dist.unina.it/proc/2011/ICWE13/start/papers/633_8page_shaping_of_bridge_box_girders_to_avoid_vortex_shedding_response.pdf, 2011-10-28. [15]. W. Zhang, Y.J. GE, Flow field mechanism of wind induced vibration response of large span bridge influenced by guide vanes, China Journal of Highway and Transport, 2009, 22(3): 52-57 (in Chinese)♦


http://www.sciencedirect.com/science/journal/01676105

http://chinesesites.library.ingentaconnect.com/content/iabse/sei;jsessionid=2roplhg3nh3lw.alexandra

http://chinesesites.library.ingentaconnect.com/content/iabse/sei;jsessionid=2roplhg3nh3lw.alexandra

http://www.dist.unina.it/proc/2011/ICWE13/start/papers/502

http://dlib.edu.cnki.net/KNS50/Navi/Bridge.aspx?LinkType=BaseLink&DBCode=cjfd&TableName=cjfdbaseinfo&Field=BaseID&Value=QLJS&NaviLink=%e6%a1%a5%e6%a2%81%e5%bb%ba%e8%ae%be

APPLICATION OF VIRTUAL SIMULATORS FOR TRAINING STUDENTS IN THE FIELD OF CHEMICAL ENGINEERING AND

PROFESSIONAL IMPROVEMENT OF PETROCHEMICAL ENTERPRISES PERSONNEL

KRASNYNSKIY M.N NIKOLAEV A.B OSTROUKH A.V State Technical University – MADI, Moscow, Russia

Summary: The structure and contents of training and control system formed on the basis of LabVIEW programming environment are presented. The training and control system can be used for teaching students in the field of chemical engineering and professional improvement of petrochemical enterprises personnel. This work was supported by the Government of the Russian Federation (Russian Ministry of Education) as part of the project under the Contract 13.G25.31.0064 on October 22, 2010.

Key word: Chemical technology, informational system, virtual simulator, SCADA, professional improvement, distant education.

Modern petrochemical industry actively uses automated control systems. These systems

not only help improving the product quality, but also provide convenient and simple tools to monitor and manage technological processes and prevent possible extraordinary situations. In order to use such systems, the enterprises’ personnel should be properly trained. Thus, it is necessary to create simulators, which are intended to give students and chemical enterprises employees an opportunity of practicing and improving their professional skills in the field of using the above systems.

Today there is plenty of software intended for training the personnel of industrial enterprises-simulators, programs for testing and so on. Introduction of such software packages at the enterprise raises the quality of personnel training and contributes to the formation of skills. It should be noted that computer-aided training involves the use of visual methods; besides, it is very convenient and easy to use. Simulators and training programs became very popular in chemical and power industries since the employees at these kinds of enterprises quite often use remote controls in their work; these operations can be easily performed by means of programs-simulators.

Besides, application of programs-simulators seems to be quite promising in training students at higher educational institutions in the petrochemical industry. It is very important for students to have both theoretical and practical knowledge which will enable them to efficiently


handle professional issues. This can be achieved through application of systems simulating the work of particular technological lines. Thus, two goals are achieved: on the one hand, students have deeper understanding of the studied material as representation of devices performance in dynamics is frequently more informative than text descriptions and static illustrative materials; on the other hand, students have an opportunity to get practical skills required for their further work, without the necessity of using real equipment which is quite often unavailable. It is obvious, that in such cases the use of simulators is, practically, a unique way to give students the necessary knowledge and skills.

The department “Computer-aided design of the process equipment” of Tambov State Technical University is engaged in the development of the automated system for control and training of the students of the specialization “Flexible automated systems in technology of machines and devices of chemical production” and the personnel of “Pigment” plc, Tambov. The given system is developed on the basis of LabVIEW programming media made by the company National Instruments.

The developed system includes the following components:

1. The virtual simulator of a workstation of the operator, controlling the work of pigment -making device.

2. The module of trainee testing, providing the knowledge check of simulated technological processes and chemical technologies as a whole.

3. The help system including the description of simulated production technology.

The training module represents a set of virtual tools created in LabVIEW system. It consists of two basic components: a simulator intended for practicing actions in case of emergency, and a simulator imitating the regular work of the technological circuit.

The simulator “PLAS-T” (fig 1) represents the virtual tool aimed at improving the personnel actions in case of emergency, created on the basis of the existing Plan of Emergency Localization (PEL) of the plant specialized in production of monometylaniline at “Pigment” plc, Tambov. It represents a complex including the forward panel, the panel of preliminary adjustments, subsystems of answers’ processing and those of testing results output.

The basic features of the system “PLAS-T” are:

- Conformity of the skills formed via the simulator, to the skills obtained through working life (it is ensured by the fact that the simulator is based on the existing PEL which is the basic document regulating the behavior of people in charge if the emergency occurs) and, thus, it secures the identity between the actions which operator is obliged to take in case of real failure and those which he does while being trained on the simulator;

- Inhibition of the skills giving a negative effect when put into real conditions as the trainee learns about the mistake or the wrong answer thus helping them to avoid a similar mistake in further testing;


Fig 1. Main panel of simulator “PLAS-T”

- An opportunity to vary testing conditions; it is ensured by two time modes of testing and the opportunity to choose a stage of technological process, its conditions and type of emergency;

- Registration of the results required for further analysis; these are presented inside the program in the form of a chart showing correct and incorrect answers, and as an external file containing complete information on testing;

- Methodical purposefulness of the simulator covering all possible options of emergency occurrence and development as stipulated in PEL and a set of the exercises for students.

A number of training modules forming “The complex of virtual simulators for chemical technology systems operators” have been developed by the students of the Tambov State Technical University on the basis of the technique developed during the creation of the described training modules, which you can find at http://www.170514.tstu.ru/tren/index.html (fig 2) is developed.

The testing module is created on the basis of “Knowledge control system Knost 1.0.4.” (www.scorp.ru), designed for the creation of electronic tests, testing and viewing of its results.

With the help of administrator module it is possible to create tests and carry out the remote testing on a local network. The test created in the system Knost, represents a set of questions and several answer options to each question. Each question may have an image in .bmp format, stored in a test file. Besides, each question has the so-called maximum point received for the correct answer. Thus, it is possible to identify the complexity of the question and its effect on the total score.


Fig 2. A web-page of the complex of virtual simulators for chemical technology systems operators

The set of questions together with the answer options covering both general knowledge of chemical production technology and the particular production process presented in the module are stored in the test file. When the test is over, the file is created; its name contains the ID of the student who has completed the test, and the title of the test itself. This file can be viewed only with the help of the program «Knowledge Control: Administrator». The score file contains the following information: the title of the test and the student’s surname, the number of questions, correct and wrong answers, factor of knowledge and a mark given according to a five - point scale. Besides, the examiner can look through the answers given by the student.

The result of the represented work is the complex system which contains the testing module for the high schools students and petrochemical enterprises employees intended to identify the level of knowledge of technological processes. The application of the given system makes it possible to arrange practical classes for the students, and teach them skills required for petrochemical enterprises. The other important area of application of the given system is training and retraining of the petrochemical enterprises personnel, revealing the degree of their readiness to various situations, including cases of emergency.

References [1]. Malygin E.N., Krasnyansky M.N., Karpushkin S.V., Mokrozub V.G., Borisenko A.B. New information technologies in the open engineering education. Textbook. / / M.: "Machinostroeniye", 2003. P. 90-123♦


OPTIMAL CONTROL STRATEGY FOR ENERGY SAVING IN TRAINS UNDER THE FOUR-ASPECT

FIXED AUTOBLOCK SYSTEM

QIHENG LU

XIAOYUN FENG Southwest Jiaotong University, Chengdu 610031, China

Summary: This paper deals with both the leading train and the following train in a train tracking under a four-aspect fixed autoblock system in order to study the optimum operating strategy for energy saving. After analyzing the working principle of the four-aspect fixed autoblock system, an energy-saving control model is created based on the dynamics equation of the trains. In addition to safety, energy consumption and time error are the main concerns of the model. Based on this model, dynamic speed constraints of the following train are proposed, defined by the leading train dynamically. At the same time, the static speed constraints defined by the line conditions are also taken into account. The parallel genetic algorithm is used to search the optimum operating strategy. In order to simplify the solving process, the external punishment function is adopted to transform this problem with constraints to the one without constraints. By using the real number coding and the strategy of dividing ramps into three parts, the convergence of GA is accelerated and the length of chromosomes is shortened. The simulation result from a four-aspect fixed autoblock system simulation platform shows that the method can reduce the energy consumption effectively in the premise of ensuring safety and punctuality.

Key words: Leading train; following train; four-aspect fixed autoblock system; optimal control strategy of energy-saving; train tracking; dynamic speed constraints; genetic algorithm.

I. INTRODUCTION

Railway transportation departments, depending on energy heavily, have a responsibility to save energy. Normally, train diagrams allow drivers to select different operating strategies, which correspond to different energy consumption.

A number of studies have been conducted to optimize train operating strategies for saving energy. With the help of genetic algorithm (GA), and using the control of coasting position, Chang and Sim [1] optimized the operating strategy of subway trains. In [2]-[5], the problem was solved by using the tool of K-T condition in optimal control theory, and the optimal control model for saving energy was built based on train control notch. By analyzing the structure of a typical section, Jin et al. [6] proposed an optimization method of train operation for saving energy based on GA. Then Fu [7] put forward a strategy of operating a following train for saving energy based on the cellular automaton model.


However, [1]-[6] all focused on a single running train; [7] aimed at the following train in a train tracking, and did not involve the leading train. In fact, every train running on a real line is interfered not only by the line conditions, but also by its leading train, whose interference is represented by the control of the signal system. Unlike that of the line conditions, the interference from the leading train is dynamic. Compared with the studies on a single train, the research targeted at both of the leading train and following train is more complicated. With the help of GA, this paper deals with the leading and the following trains in a train tracking and an optimum operating strategy is proposed for the two trains, based on the optimal operating model of trains for saving energy.

II. FOUR-ASPECT FIXED AUTOBLOCK SYSTEM

The working principle of the four-aspect fixed autoblock system is illustrated in fig. 1, with two features. One is that the signals have four aspects, which can predict the block status of three blocks forward. The other is that there are three speed grades and the braking distance from the specified speed to 0 requires two blocks.

Following train Signal

Braking curve

Leading train

Fig 1. Working principle of the four-aspect fixed autoblock system

III. MODEL

3.1. Description of the problem

Under the control of the signal system, the leading train and the following train depart in turn. After finishing their trip, they all arrive at the destination and stop. The energy consumption of the leading train and the following train are Q1 and Q2, respectively. Let f(Xi) = Qi (i = 1,2), and then the object function is demonstrated as

1 2min( ( ) ( ))f f+X X (1)

The control variables of the function f(Xi) are the train control notch and the train position when its control notch is changed. Because the control notch could be changed many times during the whole trip, the variable Xi in (1) should be a two-dimensional matrix as

(2) 1 2T1 2

1 2

, , ,( , , , )

, , ,n

i nn

l l lp p p

⎛= = ⎜

⎝ ⎠X x x x

⎞⎟

T2

Where xj (j=1,2,…,n) is the element of matrix Xi, which include the control notch lj and the train position pj. Thus, (1) is transformed into

(3) T1min( ( ) ( ))f f+X X


3.2. Dynamic equations of the trains

The running trains are affected by the traction, resistances, and braking forces, which result in different trip time and energy consumption. The dynamic equations of the trains are as follows:

( ) ( ) ( )( )

1 000v v sf r p

a− − ×

=9.8

d

(4)

(5) ( )2

vr a bv cv= + +

( )k d h hB B B K Bϕ= + = +∑ ∑ (6)

( )2 0

( ) ( )0 0d

60T T w p p

T y t

U I I tQ Q t dt

⎡ ⎤+ Δ⎣ ⎦= =∑

∫ ∫ (7)

Where a is acceleration, in m/s2; f(v), r(v) and p(s) are unit driving force, basic resistance and gradient resistance, respectively, in N/kN; a, b and c are the coefficients of basic resistance; B is the common brake force of the train; Bk is air brake force and Bd electric brake force, in kN;

hϕ is the conversion friction coefficient; ∑Kh is the total conversion brake shoes pressure of the train, in kN; QT is the total energy consumed by the train, in kW·h; is the energy consumed

by the train at time t, in s; T is the trip time, in s; U( )y tQ

W is the voltage of train pantograph, in kV; IP0 is the active current of the locomotive devices, in A; and, IP2 is the average active current of the locomotive under partial load, in A.

3.3. Constraints

3.3.1. Boundary constraints

The boundary constraints of trip time and speed are.

(0) ( )

(0) ( )

0, ,

0, 0,X

X

t t T

V V

= = ⎫⎪⎬= = ⎪⎭

(8)

Where t(0) and V(0) are, respectively, the time and speed of the train when departing; and t(X) and V(X) are, respectively, the time and speed of the train arriving at the destination.

3.3.2. Static speed constraints

The static speed constraints are defined by the line conditions, including the curve speed limits, the tunnel speed limits, the turnout speed limits, etc., demonstrated as

(9)

max_ 1 1

max_ 2 1 2max_

max_ 1

, 0 ,

, ,

, ,

s

ss

sn n

V s S

V S s SV

V S s−

≤ <⎧⎪ ≤ <⎪= ⎨⎪⎪ ≤ <⎩ X


Where Vmax_si is the static speed limits of the section i (i=1,2,…,n), in km/h; Si is the border position of section i, in m; and, s is the train position, in m.

3.3.3. Dynamic speed constraints of the following train

The dynamic speed constraints of the following train are caused by the leading train, whose position restricts the speed of the following train for safety. The constraints are closely related to the positions of the leading train and the following train, their trip time, and the signal system. Dynamic speed constraints are very important to the safety of the train tracking. After analyzing the working principle of the four-aspect autoblock system, we formulate the dynamic speed constraints of the following train as follows:

(10)

max_ 1 1 0 1

max_ 2 2 1 2max_

max_ 1

, ,

, ,

, ,

d s

dd

dn n n s n

V s S t t t

V s S t t tV

V s S t t−

≥ ≤ <⎧⎪

≥ ≤ <⎪= ⎨⎪⎪ ≥ ≤⎩

,

,s

t<

Where Vmax_di is the max safe speed of the following train in section i during the period from ti-1 to ti, and ti is the time section border of dynamic speed constraints, which is defined dynamically by the leading train position.

When the leading train passes every signal, the data of the dynamic constraints will be recalculated according to the working principle of the four-aspect autoblock system.

3.3.4. Constraints of shifting the train operating state

There are three operating states for a train, i.e., motoring, coasting, and braking. The shifting rule is listed in tab 1 [8].

Tab 1. Principle of shifting train operating state

Operating states to be shifted Current operating state Motoring Coasting Braking

Motoring

Coasting

Braking

: need not shfit; : can shfit; : cannot shfit

3.4. Object function

This problem is difficult to solve with these constraints mentioned above. In order to simplify the solving process, we introduce an external penalty function method to transform this problem to another without constraints. The object function is demonstrated as follows:

1 2min min( )E E E= + (11)


(12) 21 ( ) ( ) (0

d ( ) dT

y t X sE Q t t T loadα β= + − +∫ )0

Xs∫

∫ d d ,s t (13) 22 ( ) ( )0

d ( )T

y t xE Q t α t T= + − + ( ) ( , )0 0 0d

X X T

sβ load s γ load t s+∫ ∫ ∫

( ) max_( )

( , ) max_( , )

, ,

0, else,

, ,

0, else.

s ss

s t ds t

V Vload

V Vload

δ

η

> ⎫⎧= ⎪⎨

⎩ ⎪⎬

>⎧ ⎪= ⎨ ⎪⎩ ⎭

(14)

Where E1 is the object functions of the leading train, E2 is the object functions of the following train, load(s) is the penalty function of the static speed constraints, load(s,t) is the penalty function of the dynamic speed constraints, and α, β and γ are the factors of the trip time error, static overspeed and dynamic overspeed.

IV. GENETIC ALGORITHM

The problem under study is a nonlinear one; thus, we use GA to solve it.

4.1. Chromosomes

4.1.1. Structure of chromosomes

As mentioned above, the control variables in (3) are the train control notch and the corresponding train position, and they must be included in the genes of the chromosomes. The structure of chromosomes is illustrated in Fig. 2, where l1i and l2i are the control notches of the leading train and the following train, respectively, which both include motoring (integer, greater than 0), coasting (0) and braking (−1); p1i and p2i are the positions of the two trains where their control notches are changed.

l11p11 l12p12 … l1np1n l21p21 l22p22 … l2np2n

Fig 2. Structure of chromosomes

The length of chromosomes should be as short as possible for improving the searching speed and convergence of GA. The strategy of dividing ramps into three parts [9] is introduced here. That is, a long ramp can be divided into three parts. The train adopts the same operating mode as the previous ramp in the first part. At the beginning of the second part, the train changes the operating mode to fit the gradient of the second part. In the third part, the train changes the operating mode again to prepare for the next ramp. The real number coding strategy is also applied for the same purpose.

4.1.2. Fitness function


The chromosome is evaluated by its fitness value (must be positive) in GA. The larger the fitness of a chromosome is, the more likely the chromosome is selected into the next generation. From (11), the following fit-ness function is derived:

1 21 ( ) ,if E E= + (15)

Where fi is the fitness of the chromosome i. The fitness function guarantees two necessary conditions for the fitness. First, it is positive. Second, the less energy the trains consumes, the bigger the fitness of the chromosome.

4.2. Operators of GA

4.2.1. Selection operator

Roulette wheel selection based on elitism strategy is adopted in this paper. The probability of each chromosome in every population is calculated by (16). Refs. [10-11] demonstrates that the GA with elitism strategy is convergent in the global searching scope with the Markov chain:

sum

1

ii N

jj

if fpff

=

= =

∑ (16)

Where fsum is the total fitness of the population, and pi is the selection probability of chromosome i.

4.2.2. Crossover operator

According to the characteristics of the problem under study, the two-point parallel crossover operator is adopted. That is, a pair of parent chromosomes selected randomly is cut off at two positions. One is located in the leading train genes, and the other in the following train genes. Then the chromosome sections are exchanged and recombined to create the new offspring chromosomes.

4.2.3. Mutation operator

A mutation operator based on probability is used in this article. GA mutates the chromosomes randomly based on the preset mutation probability.

4.3. Parameters of GA

The parameters of GA are listed in tab 2.

Tab 2. Parameters of GA

Genera-tions Popula-tions Chromo-somes Probability of crossover

Probability of variation

100 1 50 0.95 0.05

V. SIMULATION

In order to verify the effectiveness of the proposed algorithm, a simulation model was


created with 1 SS8 locomotive and 18 cars. The calculation process and the parameters used here were all in conformity to Train Traction Calculation Regulation [12]. The line parameters are listed in tab 3.

Tab 3. Parameters of the line

Name Value

Departure stop Guangzhou

Destination Guangzhoudong

Length 8 510 m

Standard trip time 10 m

Signal system Four-aspect fixed automatic block signal system

The profile of the line is illustrated in fig 3, where a speed-distance curve describes an excellent driver’s operating strategy in a trip from Guangzhou to Guangzhoudong. The total energy consumption was 262.09 kW·h, and the running time was 559 s [8].

The data of braking process is calculated and the ramps of the line are simplified before the beginning of the GA according to Train Traction Calculation Regulation [12].

The data of braking process is calculated and the ramps of the line are simplified before the beginning of the GA according to Train Traction Calculation Regulation [12].

Through iterative calculation of GA, the speed-distance relation of the leading train and the following train are shown in fig 4 and fig 5, respectively.

The energy consumption of the leading train is 206.84 kW·h and the trip time is 576 s. The corresponding data of the following train are 201.79 kW·h and 562 s. Compared with the data of [8] (single train not in a train tracking, 223.086 kW·h and 603 s), the energy consumption of the leading train decreased 7.3% and the trip time decreased 4.5%, while the corresponding data of the following train are 9.5% and 6.8%.

The average fitness of all the generations of GA is illustrated in fig. 6. One can see that the maximum fitness is achieved at the 65th generation.

VI. CONCLUSIONS

(1) The static speed constraints and the dynamic speed constraints guarantee the safety of trains while running in a train tracking under the four-aspect autoblock system.

(2) It is feasible to optimize the operating strategy of the leading train and the following train with the structure of chromosomes and parallel crossover (two points) proposed in this paper.

(3) Using the real number coding and the strategy of dividing ramps into three parts, the length of chromosomes is shortened and the speed of convergence is improved.


(4) The simulation result shows that the proposed method can effectively lower the energy consumption of the two trains in a following operating in the premise of ensuring safety and punctuality.

Guangzhou Guangzhoudong

Fig 3. Profile of the line from Guangzhong to Guangzhoudong and trajectory of the train driven by an excellent driver

0 2,000 4,000 6,000 8,000 10,0000

20

40

60

80

Sped

of t

he le

adin

g tra

in (k

m/h

)

Position of the leading train (m) Fig 4. Trajectory of the leading train

0 2,000 4,000 6,000 8,000 10,0000

20

40

60

80

Sped

of t

he le

adin

g tra

in (k

m/h

)

Position of the leading train (m) Fig 5. Trajectory of the following train


0 20 40 60 80 10.08

0.10

0.12

0.14

0.16

0.18

0.20

00

Ave

rage

fitn

ess

Generation number Fig 6. Relation of GA generation and average fitness

ACKNOWLEDGEMENTS

This paper was supported by the National Science & Technology Pillar Program during the Eleventh Five-Year Plan Period of China (No.2009BAG12A05).

References

[1]. C.S. Chang, S.S. SIM, Optimizing train movements through coast control using genetic algorithm, IEEE Proc-Electr. Power Appl., 1997, 144(1): 65-73.

[2]. J.X. Cheng, P. Howlett, Application of critical velocities to the minimization of fuel consumption in the control of trains, Automatica, 1992, 28(1): 165-169.

[3]. J.X. Cheng, P. Howlett, A note on the calculation of optimal strategies for the minimization of fuel consumption in the control of trains, IEEE Trans. on Automatic Control, 1993, 38(11): 1730-1734.

[4]. P. Howlett, Optimal strategies for the control of a train, Automatica, 1996, 32(4): 519-532.

[5]. J.X. Cheng, J.S. Cheng, J. Song, et al., Algorithms on optimal driving strategies for train control problem, In: Proceedings of the 3rd World Congress on Intelligent Control and Automation, Hefei: IEEE Press, 2000: 3523-3527.

[6]. W.D. Jin, Z.L. Wang, C.W. Li, et al., Study on optimization method of train operation for saving energy, Journal of the China Railway Society, 1997, 19(6): 58-62 (in Chinese).

[7]. Y.P. Fu, Research on modeling and simulations of train tracking operation and saving energy optimization [Dissertation], Beijing: Beijing Jiaotong University, 2009 (in Chinese).

[8]. Q. He, Train optimized control based on genetic algorithm and fuzzy expert system [Dissertation], Chengdu: Southwest Jiaotong University, 2006 (in Chinese).

[9]. Y.S. Li, Z.S. Hou, Study on energy-saving control for train based on genetic algorithm, Journal of System Simulation, 2007, 19(2): 1-4 (in Chinese).

[10]. Y. Shi, S.L. Yu, Real-coded crossover operator and improved real-coded genetic algorithm, Journal of Nanjing University of Posts and Telecommunications (Natural Science), 2002, 22(2): 42-45 (in Chinese).

[11]. D. Lin, M.Q. Li, J.S. Kou, On the convergence of real-coded genetic algorithms, Journal of Computer Research and Development, 2000, 37(11): 1322-1326 (in Chinese).

[12]. The Ministry of Railways of The People’s Republic of China, TB/T 1407-1998 Train traction calculation regulation, Beijing: China Railway Publishing House, 1998 (in Chinese)♦


http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=Authors:.QT.Cheng,%20J.X..QT.&newsearch=partialPref

http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=Authors:.QT.Cheng,%20J.S..QT.&newsearch=partialPref

http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=Authors:.QT.Song,%20J..QT.&newsearch=partialPref

http://c.wanfangdata.com.cn/periodical-jsjyjyfz.aspx

http://c.wanfangdata.com.cn/periodical-jsjyjyfz.aspx

Y THE ORIENTED APPLICATION OF WEATHERING STEEL FOR BRIDGE IN VIETNAM

PROF. DR. NGUYEN VIET TRUNG DR. TRAN VIET HUNG University of Transport and Communications, Vietnam

Summary: Steel bridges are very sensitive with environmental conditions, thus it is necessary to protect rust of bridge structure. However, the cost of the anti-rust and maintenance is especially high. Recently, new steel technologies have rapidly developed and created many new steel materials by steelmakers such as high-performance steel, weathering steel. Weathering steel is available anti-rust by itself and low maintenance cost. The exposure test specimens were attached on Cho Thương Bridge. The results of the test based on weathering steel (Japanese Industrial Standard, JIS-SMA-W) has showed good corrosion protection. Thus, the oriented application of weathering steel on the bridges in Vietnam, where have a long coastline of more than 3,200 km, is studied in this paper.

Key word: Bridge, steel bridge, weathering steel, high performance steel

I. INTRODUCTION

Steel bridges without painting or other protective coatings have been used widely for the sake of environment protection and low cost (Mathay, W.L., 1993). Recently, weathering steel bridge has received increased attention because correct applications of weathering steels to infrastructures are beneficial for reducing maintenance cost. Weathering steel use commonly in Unites States, named Cor-Ten™, is a group of steel alloys that develop a stable oxidation requiring no additional coating. Almost weathering steel is ASTM A 588, also known as Cor-Ten™ B, an "improved" alloy. Material is available in rolled shapes and plate conforming to ASTM A 709, which can be fabricated into various structural shapes, and in coiled sheet conforming to ASTM A 606, which is formed into roofing and siding panels.

In Japan, the conventional weathering steel specified as Japan Industrial Standard G 3114 SMA (JIS-SMA weathering steel) and advanced weathering steels from three Japanese steelmakers, are available commercially. When the weathering steel bridges construct, it is important to judge whether the weathering steels applied to the structure are suited to the local environment. The Japanese Roads and Bridges Policy Manual states that JIS-SMA weathering steel can be used without rust controlling surface treatment for bridges in regions where the deposition rate of air-bornsalt is less than 0.05 mdd (mg/dm2/day) because corrosion loss is related to deposition rate of air-born salt (Ohya, M. et al., 2009).

A rust layer formed on a low-alloy steel surface is generally considered to be responsible for protection of the steels against corrosives in atmospheric environment. Weathering steel possesses high corrosion resistance, approximately twice of that of carbon steels, and therefore


has been widely used as a structural material in an atmospheric corrosion environment. The protective ability of the rust layer emerges, the corrosion rate of the weathering steel is not considerably low and initial un-protective rust contaminates surfaces of itself and other environing materials. The protective rust layer cannot form in a coastal environment where the amount of air-borne sea-salt particle is relatively high. These are significant problems for reducing maintenance task for weathering steel structures (Yamashita et al., 2002). However, the standard to be applied to advanced weathering steels is not stipulated clearly. Corrosiveness of atmosphere differs by location, so a corrosion prediction method for both conventional and advanced weathering steels is needed at the planning and/or design stage to ensure selection of materials that will enhance structural durability (Ohya, M. et al., 2009).

Almost steel bridges in Vietnam have been constructed long time ago, have not been used new steel materials and have been down graded serious. Thus, the new steel as weather-resistant steel is necessary to research for applying in Viet Nam. Three railway bridges in the North-South railway system (N-S railway) have been constructed for this purposing. The results show an application capacity of new steel materials is good on Vietnam’s condition. Overview of the development situation of the steel bridge and the ability and potential application of new steel materials in Vietnam is presented in this paper.

II. THE SITUATION AND DEVELOPMENT OF THE STEEL BRIDGE IN VIETNAM

Viet Nam has along a coastline of 3,260 km across the territories of 24 provinces and cities, which including 127 urbans and rural districts, 21 towns and 6 large-cities (Ha Long, Vinh, Hue, Quy Nhon, Nha Trang, Vung Tau). The national highway system runs nearly a coast, and a railway system runs cross sparsely populated areas with a lot of steel bridges. The rust protection of the structural steel bridges are necessary maintain within a short period, however, it have difficult with cost and technology. Therefore, a lot of steel bridges are being seriously rust in Vietnam (shown in Fig. 1). Steel bridges will be built in the new transport systems. Thus, it is necessary to research the applied orientation of new steel technology to bridge in the future.

Fig. 2 shows the concrete bridges of 89% and the steel bridges of 11% in Vietnam). However, the steel bridges have constructed long time ago. Almost steel bridges have been constructed since more than 20 years ago, such as Thang Long truss bridge was completed since 1980, Chuong Duong truss bridge was completed since 1986, etc. (shown in fig. 3). Steel materials used mostly carbon steel and low alloy steel as shown in table 1. Recently, few steel bridges have been constructed in Viet Nam such as the Ho truss bridge was completed in 2000; the Binh cable stayed bridge with main span of 250 m was completed in 2005. This bridge is the first cable stayed bridge used steel girder in Vietnam with 17 spans of continuous composite girder, a main span of 260m, effective width of road 22.5 m (4 lanes + 2 sidewalks) and total length of 1.280 m. Steel girder and steel anchor box used steel of 6.700 tons including SM570, SM520C, SM490Y, SM490 and SM400A following to Japan Standard-JIS (Matsuno et al., 2006); the Thuan Phuoc bridge is the first suspension bridge used steel box girder in Vietnam with main span of 405 m was completed in 2009; the Rao II and the Nhat Tan cable stayed bridges are constructing in Vietnam.


Fig. 1. Corrosion of steel bridge in Vietnam Steel bridge

11%

Concrete bridge 89%

Fig. 2. Percent of steel bridge in Vietnam (Source: VRA)

Binh bridge

Thang Long bridge

Chuong Duong bridge

Fig. 3. Typical steel bridge in Vietnam

Vietnamese specifications for bridge design and material standards have been changed as 22TCN 18-79, (referred to Russia standard) has been replaced by 22TCN 272-05 (referred to ASSHTO LRFD 1998). The new specifications for bridge design (22TCN 272-05) has been officially used on 2005, and materials standard based on ASTM standard. Currently, Vietnam has no design standard requirements for high-performance steel materials as weathering steel.

Table 1. Characteristics of steel structures were applied in Vietnam

Carbon steel following to Vietnam standard (TCVN 5709: 1993) Yield strength, fy (N/mm2) with thickness, t (mm) Symbol t ≤ 20 20 ≤ t ≤ 40 40 ≤ t ≤ 100

CCT34 CCT38 CCT42

220 240 260

210 230 250

200 220 240

Low alloy steel following to Vietnam standard (TCXDVN 338: 2005) Yield strength, fy (N/mm2) with thickness, t (mm) Symbol t ≤ 20 20 ≤ t ≤ 40 40 ≤ t ≤ 100

09Mn2 14Mn2 16MnSi

09Mn2Si 10Mn2Si1

10CrSiNiCu

310 340 320 330 360

400*

300 330 300 310 350

400*

- -

290 290 340

400* Note: * γM = 1.1 with maximum thickness of 40 mm


III. DEVELOPMENT AND CHARACTERISTICS OF WEATHERING STEEL

Weathering steel used for bridges is low alloy steel containing small amounts of corrosion resistant elements such as copper (Cu), nickel (Ni), and chromium (Cr) (Table 2). During its use without painting, compact and protective rust having good adhesion forms on steel surfaces, which suppresses further progress of corrosion to a sufficiently low level. A characteristic of weathering steel bridge is the rusty layer on weather resistant steel restrains corrosion. The usage of this steel helps to reduce the total maintenance cost. A consideration of bridge lifetime when designing, the life cycle cost can be reduced.

Table 2. Ni-advanced weathering steels Material Components

Ni-advanced weathering steel

3.0% Ni-Cu 2.7% Ni-Cu-Ti 2.5% Ni-low carbon 1.5% Ni-Mo 1.2% Ni-Cu 1.0% Ni-Cu-Ti

JIS Weathering steel 0.3% Cu – 0.5% Cr

Fig. 4. Stable oxidant layer (Source: Japan bridge associate)

The application of weathering steel to steel bridges began in the 1960s in Japan. The protective rust does not always form as designed owing mainly to air-borne sea-salt, resulting in lamellar exfoliation of rust layers. In view of the situation, unpainted use of the weathering steel for bridges is recommended, without requiring actual measurement of the air-borne sea-salt, only at locations where the deposition of airborne sea salt is 0.05 mdd (mg/dm2/day) or less, measured by the method described in JIS Z2381 (Japan Road Association, 2002). This steel possesses a unique property of suppressing the development of corrosion by a layer of densely-formed fine rust on its own steel surface: the corrosion rate gradually reduces to the level that causes virtually no damage from an engineering viewpoint, as the layer of the rust grows. Thus, the coating is not required in the weathering steel bridge, the cost of which therefore can be much lower than that of a conventional steel bridge. Fig. 4 shows characteristic of 3%-Ni weathering steel in Japan.


V. ORIENTED APPLICATIONS OF WEATHERING STEEL BRIDGE IN VIETNAM: CASE STUDY OF CHO THUONG BRIDGE

5.1. Description of Cho Thuong bridge and items of the investigation

Three bridges using weathering steel on the N-S railway have been constructed in Viet Nam (Trung and Hung, 2007). It located between Ha Noi capital and Ho Chi Minh city (fig. 5), was constructed at the 338 km from Ha Noi (about 25 km away from the sea) in Ha Tinh Province on May 2000. This is a truss bridge with unpainted weathering steel. This bridge includes 4 simple spans with a length span of 61 m and effective width 4.26 m. Truss structures used SMA400AW, SMA400AP and BP. Some locations of structural members have been painted and unpainted as above rails and under rails respectively (fig. 6). We also installed exposure test pieces, named “Button-test specimen” to investigate the rust every years. The items and methods of the investigation show in table 3.

Investigation position Cho Thuong bridge

Railway

HA TINH Province

Fig. 5. Location map of Cho Thuong bridge Fig. 6. General profile of Cho Thuong bridge

Fig. 7. Actual state of structural members

Fig. 8. Arrangement of the exposure test specimens

5.2. Environment conditions of Ha Tinh province

Cho Thuong bridge belongs to Ha Tinh province and the tropical area. Ha Tinh has two clearly climate seasons: rainy season from August to November with average rainfall over 2000 mm and dry season from December to July of the next year. The dry season is very hot with hot-wind stream from Laos.


Table 3.The items and methods of the investigation

No. Items Methods

1 Appearance Rust condition

Visual investigation, photos Comparison with the standard photos of rust

2 Measurement of rust thickness Electromagnetic instrument to measure thickness

3 Thickness loss of steel dueto corrosion after 1,2,3 years by Button-test

Attach exposed Button-tests pecimens to the bridge Next time: Take them off after 1,2,3 years Measure the thickness loss and compare the corrosion of the specimens with it of the bridge itself

4

Environmental data: flying salt, temperature, humidity, wind velocity, sulfur oxide, for a year

Check exiting data from the meteorological authority or others

Table 4. Rating of air-born salt in Vietnam (QCVN 02: 2009/BXD)

Zones S mmd (mg/m2/day) Regional characteristics

1 S >4 Marine areas and islands

2 4≥S>2 Coastal

3 2≥S>0.5 Delta

4 0.5≥S>0.4 Region far sea and midland

5 S≤0.4 Midland far sea and mountain

Ha Tinh lies in the tropical and monsoon belt, gives a typical tropical climate in southern

and a cold and dry winter in northern. Annual average rainfall from 2,500 mm to 2,650 mm

occurred in rainy season within heavily rainy period of the late August to mid-November.

Especially, the highest rainfall of 3,000 mm occurred in this period. The lower rainfall from

1,000 to 1,500 mm also occurred in the other areas. Dry season occurs from December to July

of the next year with an intensively sunny period of dry, hot and heavily evaporative South-

West wind blowing from Laos. The temperature from April to mid-August is very hot with

average temperature of 290 C, reaching 40-410 C on some days. The temperature from October

to March next year is cold with the temperature down to 6-70 C (http://ngoaivuhatinh.gov.vn).

The detailed data referred to the QCVN 02: 2009/BXD (fig. 8). The air-born salt on almost

coastal and in-land region in Vietnam is less than 4 mmd (= 0.04 mdd(mg/dm2/day)) (table 4).


http://ngoaivuhatinh.gov.vn/

Jan Feb Mar Apr May June July Aug Sep Oct Nov DecWind speed (m/s) 1.5 1.4 1.3 1.3 1.4 1.4 1.6 1.4 1.4 1.9 1.8 1.7Temperature (C‐degree) 17.6 18.1 20.7 24.3 27.8 29.3 29.7 28.7 26.8 24.3 21.4 18.6Evaporate (mm) 40 43 57 72 87 115 110 98 65 50 48 42Moisture (%) 91 92.9 91.8 88.3 81.7 76.8 73.9 79.7 86.5 88.7 88.5 88.2Sunshine (hours) 77 50 76 137 219 206 233 193 161 135 95 82Rainfall (mm) 97 64 54 74 143 144 112 225 532 765 319 162

1

10

100

1000

Fig. 9. Statistic average environment conditions in Ha Tinh province (QCVN 02: 2009/BXD)

Rust Grade 4 Average thickness of rust (μm) 90.1

Rust Grade 3 Average thickness of rust (μm) 124



Fig. 10. Rust investigation in span P1~P2 (unpainted) for right side of truss (left fig.) and left side of truss (right fig.)

5.3. Appearance of rust

The rust investigation of the unpainted steel has shown in Figs. 10 and 11. The rust grade of 4 is estimated at most members of this bridge. The rust grade of 3 is partially shown on the surface of the diagonals, where contact to strong wind blows. The rust grade estimation of 3 and 4 show the rust state of this bridge is good condition. The average rust thickness of each surface is around 80 to 140μm and 110 to 190μm for the grade of 4 and 3, respectively. The rust thickness does not show the thickness loss of steel as rust expands.


According to the data measured in Japan, the thickness loss is about 30% of the rust thickness. Thus, it is thought to be around 20 to 40μm and 30 to 60μm in case grade of 4 and 3, respectively. This is very good condition. As regards the condition of the painted surface of the lower members was also good (fig. 11). As the results, the unpainted and painted weathering steel of Cho Thuong bridge is estimated to good condition.

Rust Grade 4

Average thickness of rust (μm) 82.8


Fig. 11. Rust investigation in span P3~A2 (unpainted elements)

Rust Grade Good Average thickness of rust (μm) 371

Rust Grade Good Average thickness of rust (μm) 597

Fig. 12. Rust investigation in span P1~P2 (painted elements)

5.4. Corrosion estimation of weathering steel

This bridge was constructed since 12 years ago. Thus, the corrosion estimation of this bridge is limiting because it is necessary long term estimation on the existing bridge and exposure button test using attachable small test pieces made of weathering steel. We used short-term exposure testing method to predict long-term corrosion losses that could occur on the weathering steel bridges. In this study, the corrosion prediction is performed with limited investigate results in recent years.

Method of predicting loss of thickness of weathering steel due to corrosion is used in this study (Kihira et al., 2005; Fijii et al., 2008). The penetration curves of weathering steels can be expressed as follows:

(1) BY = A X

Where X is time in years, Y is the penetration (mm), A is the first year corrosion loss (mm), and B is the index of corrosion rate diminution.

Based on the results of short-term exposure tests, the loss of steel thickness due to corrosion of SMA weathering steel in the first year of the test (X = 1 year) is the first evaluated as the environmental corrosiveness index ASMA. The value of ASMA corresponds to the value of Y (thickness loss due to corrosion). On the other hand, the value of BSMA corresponds to the


reciprocal of the degree of decrease in corrosion rate due to an increase in the protective action of the rust. Using the short-term exposure testing results, we tried to find the value of A and to estimate the corrosion loss after 100 years by means of estimation procedure as shown in fig. 13.

The mean of the loss prediction result is fairly small of 0.32 mm when the distance from the coast is about 25 km. However, the long-term exposure test is necessary carried out to estimate the applicability of this steel type with Vietnam conditions.

After 5 years Estimated loses in 100 years Fig. 14. The Cho Thuong raiway bridge, Ha Tinh province [25 km from coastal region] – COR-TEN

ASMA = 0.019, B BSMA = 0.62, Y = 0.32 mm, (0.23-0.46), [X=100]

VI. CONCLUSION

Weathering steel has been used for many structures including bridges because it’s unique property of preventing rust by rust. Many successful instances have been witnessed with the reduction of maintenance and control cost. In this paper, we outlined applicability evaluation for weathering steel and the advantages of applying them to bridges in Vietnam. Through above analysis and investigation show applicability of weathering steel in Vietnam as follow:

- Oriented application high-performance steel, and weathering steel is very feasible in Vietnam condition with a long coast-line, 3260km, and narrow width.

- The use of ability restrains corrosion of it-self is very good in the sparsely populated areas and less maintenance conditions.

- The air-born salt is not too large (most less than 4 mmd at the coastal and inland in Vietnam) allows better application of these steel structures in Vietnam.

- The investigation of loss corrosion with SMA weathering steel are using at Cho Thuong bridge since 12 years ago shown good corrosion protection within local environments. The prediction results for loss corrosion of this steel type within 100 years also showed non-significant. However, it is necessary to observe the course by continuing to conduct exposure tests and to verify the reliability through collecting the data in long-term exposure tests at some locates and various environments.

Although the applicability large but also in future need more research on cost, construction technology and especially, a new design standard in order to creating sustainable developing in


the bridge construction in Vietnam.

References [1]. Akira, U., Hiroshi, K., Takashi, K.: 3%-Ni Weathering Steel Plate for Uncoated Bridges at High Airborne Salt Environment, Nippon steel technical report, No. 87, pp 21-23, 2003. [2]. Miki, C., Homma, K., Tominaga, T.: High strength and high performance steels and their use in bridge structures, Journal of Constructional Steel Research, Vol. 58, pp. 3–20, 2002. [3]. Eiki Y.: Application Potential of Weathering Steel Bridges, Steel construction Today & Tomorrow (JISF-JSSC), Vol. 15, pp. 11-12, 2006. [4]. Fujii, Y., Tanaka, M., Kihira, H. and Matusoka, K.: Corrosion risk management methods to realize long-term durability of weathering steel bridges, Nippon Steel Technical Report, No. 387, 2007. [5]. Kihira, H., Senuma, T., Tanaka, M., Nishioka, K., Fujii Y., and Sakata, Y.: A corrosion prediction method for weathering steels, Corrosion Science, Vol. 47, pp. 2377–2390, 2005. [6]. Japan Road Association: Design Specifications for Highway Bridges, Part II Steel Bridges. Tokyo: Japan, 2002. [7]. Japanese Society of Steel Construction, “Possibility and new technologies of weathering steel,” 2006 (in Japanese). [8]. JSSC: Potentials and New Technologies of Weathering Steel Bridges, JSSC Technical Report No. 73, 2006. [9]. Isamu, K., Kazuyuki, M., Fumimaru, K.: Minimum Maintenance Steel Plates and Their Application Technologies for Bridge - Life Cycle Cost Reduction Technologies with Environmental Safeguards for Preserving Social Infrastructure Assets, JFE Technical report, No. 5, pp. 37-44, 2005. [10]. Kihira, H., Tanabe, K., Kusunoki, T., Takezawa, H., Yasunami, H., Tanaka, M., Matsuoka, K. and Harada, Y.: Mathematical modeling to predict long-term corrosion loss to occur on weathering steels, J. Structural Eng./Earthquake Eng., JSCE, No.780/I-70, pp.71-86, 2005. [11]. Ohya, M., Takebe, M., Adachi, R., Ota, J. and Kitagawa, N.: Applicability of a New Method for Selecting Weathering Steel for Bridges, Corrosion in Marine and Saltwater Environments 3 214th ECS Meeting, Vol. 16 (43) pp. 87-99, 2009. [12]. Yamashita, M. and Uchida, H.: Recent Research and Development in Solving Atmospheric Corrosion Problems of Steel Industries in Japan, Hyperfine Interactions 139/140: 153–166, 2002. [13]. Masaaki, U.: Tokyo Bay Bridge - Long-span Bridge Employing Advanced Bridge Materials and Technologies, Steel construction Today & Tomorrow (JISF-JSSC), Vol. 18, pp. 1-6, 2007. [14]. Matsuno, K., Nakayama, M., Fujita, K., Yamamoto, Y. and Oyama, A.: Construction report for “Binhbridge” in Vietnam, IHI engineering review, Vol. 39, No. 2, pp.47-64, 2006. [15]. Miki, C., Ichikawa, A., Kusunoki, T. and Kawabata, F.: Proposal of New High Performance Steels for Bridges (BHS500, BHS700), JSCE, No. 738/I-64, pp.1-10, July 2003. [16]. Trung, N.V. and Hung, T.V: Characteristics of weathering steel and its application in steel structure construction in vietnam, Journal of Bridge and Road Engineering, Vietnam, Vol. 3, P. 25-28, 2007. [17]. QCVN 02: 2009/BXD, Vietnam Building Code - Natural Physical & Climatic Data for Construction, 2009. [18]. The Japan Iron and Steel Federation, and the Japan Bridge Association: Application of Weathering Steel to Bridges, 2003. [19]. Vietnam Ministry of Transport, Specification for bridge design - 22TCN 272-05, 2005. [20]. Mathay, W.L.: Highway structures design handbook, American Institute of Steel Construction Inc., 1993. [21]. Fujino, Y.: Steel Bridges in Japan - Current Circumstances and Future Tasks, Steel construction Today & Tomorrow (JISF-JSSC), Vol. 15, pp. 1-3, 2006. [22]. Yamaguchi, E., Nakamura, S., Hirokado, K., Morita, C., Sonoda, Y., Aso, T., Watanabe, H., Yamaguchi, K. and Iwatusbo, K.: Performance of weathering steel in bridges in Kyushu-Yamaguchi region, Doboku GakkaiRonbunshuu A, JSCE, Vol.62, No.2, pp.243-254, 2006♦


AN IMPROVED ALGORITHM FOR FLUID-STRUCTURE INTERACTION OF HIGH-SPEED TRAINS UNDER CROSSWIND

TIAN LI JIYE ZHANG WEIHUA ZHANG Traction Power State Key Laboratory, Southwest Jiaotong University, Chengdu 610031, China

Summary: Based on the train-track coupling dynamics and high-speed train aerodynamics, this paper deals with an improved algorithm for fluid-structure interaction of high-speed trains. In the algorithm, the data communication between fluid solver and structure solver is avoided by inserting the program of train-track coupling dynamics into fluid dynamics program, and the relaxation factor concerning the load boundary of the fluid-structure interface is introduced to improve the fluctuation and convergence of aerodynamic forces. With this method, the fluid-structure dynamics of a high-speed train are simulated under the condition that the velocity of crosswind is 13.8 m/s and the train speed is 350 km/h. When the relaxation factor equals 0.5, the fluctuation of aerodynamic forces is lower and its convergence is faster than in other cases. The side force and lateral displacement of the head train are compared between off-line simulation and co-simulation. Simulation results show that the fluid-structure interaction has a significant influence on the aerodynamics and attitude of the head train under crosswind conditions. In addition, the security indexes of the head train worsen after the fluid-structure interaction calculation. Therefore, the fluid-structure interaction calculation is necessary for high-speed trains.

Key words: High-speed train; fluid-structure interaction; crosswind; aerodynamics; relaxation factor.

I. INTRODUCTION

High-speed transportation is the new direction of modern railway transportation [1]. High-speed train aerodynamics and train-track coupling dynamics, the indispensable parts of high-speed transportation system, are mutually coupled and influenced. The action of aerodynamic forces will change the running attitude of the train, and consequently the running attitude will affect the flow field around train. Thus, the aerodynamic forces of the train will change, and the train system will be in a particular state of coupling vibration under this mutual feedback.

Strong crosswind seriously affects the running security of high-speed trains. The train derailments and overturns happen because of strong crosswind [2]-[5]. Many researchers [6]-[11] have analyzed the running security of trains under crosswinds. There are mainly two calculation methods of high-speed dynamics under crosswind conditions: off-line simulation method, and co-simulation method. The former is as follows: the aerodynamic forces of the train under crosswind conditions are calculated first, and then the dynamic responses of the train are


calculated with the train dynamic model, on which the static aerodynamic forces are loaded. This method neglects the change of attitude caused by the action of aerodynamic forces, and does not reflect the essence of the aerodynamics and train dynamics. In the co-simulation, the high-speed train aerodynamics and train-track dynamics are calculated alternatively during the interaction; namely, the interaction effect between the high-speed train aerodynamics and train-track dynamics is considered. Yang et al. [10] and Cui et al. [11] performed a co-simulation between high-speed train aerodynamics and train dynamics through parameter transfer and synchronization control, but they neglected the influence of track structures on the system [11]. In Refs. [10]-[11], the train dynamic solver is in a state of either calculating or waiting, which consumes too much memory and resources. In addition, the information exchange on the interface from one time step to the next time step in an alternating fashion can easily cause energy dissipation.

In this paper, a co-simulation algorithm was improved by (1) inserting the program of train-track dynamics into the program of computational fluid dynamics to avoid the data communication between fluid and structure solvers, and (2) introducing the relaxation factor to improve the information exchange on the interface between aerodynamics and train dynamics. With the improved algorithm, we conducted a co-simulation between aerodynamics and train-track dynamics for a train running at a speed of 350 km/h under a crosswind of velocity 13.8 m/s, and analyzed the fluid-structure dynamics of the high-speed train.

II. GOVERNING EQUATIONS

2.1. Governing equations of fluid dynamics

When the high-speed train is running under crosswind, its flow field can be considered as a three-dimensional transient viscous turbulent flow. When the running speed is lower than 400 km/h, the flow field around the train can be considered as an incompressible flow. The standard k-ε two-equation model is adopted and the equations of incompressible flow [9] are written as

( )( ) div( ) div ,t

ρ ρ Γ∂+ =

∂u gradφ φ + Sφ (1)

Where t is time, ρ is the air density, u is the velocity vector, φ is flow flux, S is the source term, and Γ is the diffuse coefficient.

2.2. Equations of train-track coupling dynamics

The train-track coupling dynamics mainly include vehicle dynamics, track dynamics, and wheel-rail contact. It is assumed that the body, bogies, and wheelsets are rigid and that their elastic deformations can be neglected. The track system is considered as a continuously distributed spring-damper model with a two-mass (sleeper and ballast) and three-layer (rail–sleeper–ballast-bed) structure. The equations of the train-track coupling dynamics [9] are written as


(2) ,+ + =MX CX KX F

Where M, C, and K are the mass, damp, and stiffness matrixes of train-track system, respectively; X, X , and X are the generalized displacement, velocity, and acceleration vectors of the system, respectively; and F is the generalized load vector including the rail excitation and aerodynamic forces load.

III. IMPROVED ALGORITHM FOR FLUID-STRUCTURE INTERACTION

3.1. Technique for solving train-track dynamics equation

Based on the theory of train-track coupling dynamics, the program of train-track coupling dynamics is written with FORTRAN and verified to be reliable [9].

The train-track dynamics equations are solved with the Zhai's method [12]. Introducing two integral parameters and ψ, we construct the new explicit integral format as:

2 2

n+1 n n n n-1

n+1 n n n-1

1X = X + X Δt + ( +ψ)X Δt -ψX Δt ,2

X = X + (1+φ)X Δt -φX Δt,

⎧⎪⎨⎪⎩

(3)

Where Δt is the integral interval, and subscript n means the time iteration step.

The form of (3) at time t=(n+1)Δt is

1 1 1n n n 1n+ + ++ + =MX CX KX F + (4)

Substituting (4) into (3), we can calculate . n+1X

3.2. Mesh renewing technique

The spring analogy method [12] and the re-mesh method are adopted to renew the mesh. When the spring analogy method fails, the re-mesh method is adopted. Spring analogy method is a simple but efficient method among mesh deforming methods. In this method, each edge of the grid is modeled as a linear tension spring. The spring stiffness for a given edge i-j is taken to be inversely proportional to the length of the edge as

ijij i j

1 1K = =r || r - r ||

(5)

Where rij is the distance between node i and j, and ri is the position vector of node i.

The displacements of grid points is solved by

(6) iN

ij jj

K Δr = 0∑

Where Ni is the total number of the nodes connected with node i, and Δrj is the displacement of node j. The summations are performed over all the edges of quadrilaterals with


node i as an end point, i = 1,2,…,n.

The new locations of the nodes are determined by

.i i i= + Δr r r (7)

3.3. Solution strategies

Fig 1 shows the co-simulation method of high-speed train fluid-structure interaction. The process of the method is described as follows: first, calculate the aerodynamic forces of train under crosswind until the forces reach a relatively steady state, and then the iteration between aerodynamics and train-track coupling dynamics begins. At each iteration, the transient aerodynamic forces are transferred from the fluid solver to the train dynamic model, and the responses of train dynamics are calculated with this model. Then the attitudes are transferred to the fluid solver and the aerodynamic forces are calculated under the train attitude. In this method, the state of train dynamic solver is either calculating or waiting, causing waste of memory and resources. In addition, the load boundary of aerodynamics and dynamics is very simple, which may cause energy dissipation.

Fig 2 shows an improved co-simulation method of high-speed train fluid-structure interaction. The fluid-structure solver includes the aerodynamic solver and train-track coupling dynamic solver, and the latter is inserted into the aerodynamic solver through interface. Thus, the data communication between fluid and structure solvers is avoided, and the train-track coupling dynamic solver does not have to wait when the fluid solver is calculating. The relaxation factor is introduced for renewing the boundary of aerodynamics and dynamics. The aerodynamic forces loaded to the train-track dynamics model at time i+1 is no more than the forces at time i, which is predicted by the aerodynamic forces and their velocities at time i.

It is assumed that the aerodynamic forces fi-1 and fi are calculated by the fluid solver at time i-1 and i, respectively, and the time interval is Δt. As the Δt is very small, the velocity of aerodynamic forces at time i is

1i ii

f ff

t−−

=Δ

(8)

Aerodynamics

Tr in-track

coupling dynamics

1

1

Forc

0 …

Forces

i-1

i-1

Attitude

i

i

Forces Attitude

i+1

i+1

Forces Attitude

… …

…

Fig 1. Co-simulation method of high-speed train fluid-structure interaction

Fig 2. oved co-simulation method of high-speed train fluid-structure interaction

Aerodynamics

Train-track coupling dynamics

1

1

Forces Attitude

0 … i-1 i i+1 …

i-1 i

Forces Forces Forces

i+1

Attitude Attitude Attitude


100 m

60 m

90 m

175 m

1.

Fig 3. matic of computational domain

Thus, the predicted aerodynamic force at time i+1 is

1 2i i i i if f t f f f+ −= + Δ × = − (9)

By introducing the relaxation factor λ, the predicted aerodynamic force 1if + , which is loaded to the train-track dynamic model at time i+1, is expressed as follows:

1 1(1 ) (1 ) .i i i i 1if f f f fλ λ λ λ+ += − + = + − − (10)

When λ=0, the aerodynamic force loaded to the train-track dynamic model at time i+1 equals fi. Under this condition, the method is similar to the method in [10]-[11]. IV. SIMULATION 4.1. Computational domain and boundary

In this section, we describe the fluid-structure problem of the high-speed train under crosswind. The schematic of the computational domain is shown in fig 3. The computational range is 350 m in length, 90 m in width and 60 m in height. The distance from the inlet boundary to the nose of the head train is 100 m, and the distance from the outlet boundary to the nose of the tail train is 175 m. Velocity inlet condition and traction-free condition are preset at the inlet boundary and outflow boundary, respectively. In addition, a no-slip condition and symmetry condition are specified as the train surface and top boundaries, respectively. The slip condition is adapted to the wall boundary.

The train consists of three cars, including head train, middle train and tail train, with the bulge (such as the pantograph) ignored. The iteration time steps of the fluid dynamic and the train-track dynamic are 2.0×10-3 s and 5.0×10-5 s, respectively. In this paper, calculations are carried out for a high-speed train with a speed of 350 km/h and the crosswind velocity of 13.8 m/s; that is, the combination velocity equals 98.2 m/s and the yaw angle is 8.08°. 4.2. Influence of relaxation factor

The load boundary of fluid-structure interface, with different values of the relaxation factor, may affect the energy dissipation. In this section, three different relaxation factors (λ=0.0, 0.5 and 1.0) are chosen to analyze the effect.

Fig 4 shows the side force and lift force curves calculated with different relaxation factors. The change curves of forces are almost the same, except some differences in the amplitudes of side force and lift force with different relaxation factors. The amplitude of side force becomes larger with an increase in relaxation factor. However, the opposite is true with lift force. One can see that there are fluctuations of aerodynamic forces.


0 3 6 9 12-56

-54

-52

-50

-48

-46

15

λ=0.0 λ=0.5 λ=1.0

Forc

e (k

N)

Time (s) (a) Side force

0 3 6 9 12-14

-13

-12

-11

-10

-9

-8

15

λ=0.0 λ=0.5 λ=1.0

Forc

e (k

N)

Time (s) (b) Lift force

Fig 4. Side force and lift force with different λ

In order to analyze the influences of aerodynamic forces on the relaxation factor, the average standard deviation is introduced for evaluating the fluctuation of aerodynamic forces. The average standard deviation is described as

2

2

3

1 ( )5

n

i ii

S fn

−

=

= −− ∑ ,f (11)

2

2

1 3, 4,..., 2,5

j

j jj j

f f j n+

= −

= =∑ − (12)

Where n is the total number of time steps, and if is the aerodynamic force at time i obtained by the five-point mean method.

Fig 5 shows the average standard deviation of drag force, side force, lift force, roll moment, pitch moment and yaw moment with different relaxation factors. When the relaxation factor equals 0.5, the average standard deviation of force or moment is nearly the minimum in the three cases. That means the fluctuations of aerodynamic forces are relatively smaller.

In addition, another definition of standard deviation of error is introduced for evaluating the error between the predicted aerodynamic forces and the calculated aerodynamic forces. The standard deviation of error is described as

1

2

2

1 ( ) . (13) 2

n

i ii

S f fn

−

=

= −− ∑


0

5

10

15

20

25

30

35

Lift forceSide forceDrag force

λ=0.0 λ=0.5 λ=1.0

Ave

rage

stan

dard

deb

iatio

n

(a) Forces

0

30

60

90

120

150

Yaw momentPitch momentRoll moment

λ=0.0 λ=0.5 λ=1.0

Ave

rage

stan

dard

deb

iatio

n

(b) Moments

Fig 5. Standard deviation of aerodynamic forces Fig 6 shows the standard deviation of error of drag force, side force, lift force, roll moment,

pitch moment and yaw moment with different relaxation factors. Similarly, when relaxation factor equals 0.5, the standard deviation of error of force or moment is nearly the minimum in the three cases. From Eq. (13), one knows that the predicted aerodynamic forces are more close to the calculated aerodynamic forces.

By comparing average standard deviation and standard deviation of error of the aerodynamic forces calculated with different relaxation factors, one can see that the fluctuation of aerodynamic forces is small and that the predicted aerodynamic forces are closer to the calculated aerodynamic forces when the relaxation factor equals 0.5.

0

20

40

60

80

Lift forceSide forceDrag force

λ=0.0 λ=0.5 λ=1.0

Stan

dard

dev

iatio

n of

err

or

(a) Forces

0

50

100

150

200

250

300

350

Yaw momentPitch momentRoll moment

λ=0.0 λ=0.5 λ=1.0

Stan

dard

dev

iatio

n of

err

or

(b) Moments

Fig 6. Standard deviation of error of aerodynamic forces

4.3. Fluid-structure interaction dynamics

Fig. 7 shows the pressure distribution of a high-speed train cross-section. The cross-section is 14 m away from the nose of the head train. We can see that, under the interaction effect, the pressure of the train windward side becomes larger and the pressure on the lee side becomes smaller. In addition, the pressure of the train bottom becomes larger because it has turned into lee side after the roll of the train body.

The side force comparison of the head train is shown in Fig. 8. The side force calculated with off-line simulation is a steady value; however, the one with co-simulation shows a fluctuant curve. After considering the interaction effect, the side force becomes larger than before. The side force is closely related to the pressure difference between the lee side and the windward side. According to the analysis of pressure distribution around the train cross-section, as shown in fig. 8, the pressure difference becomes larger after taking the change of train


attitude into account. Thus, the side force calculated with co-simulation is over 10% more than that with off-line simulation.

(a) Off-line simulation

(b) Co-simulation

Fig 7. The pressure distribution of cross-section

The train-track coupling dynamic responses to the aerodynamic forces are calculated. There are some differences in the dynamic responses because of the differences of forces and moments. Fig 9 shows the lateral displacement comparison of the head train. After we take the interaction effect into account, the lateral displacement of the head train toward the lee side is larger than before. However, the variations of lateral displacement are similar as the lateral displacement mainly depends on the lateral track irregulity.

Tab 1 shows the comparison of train security indexes calculated by different methods. After considering the interaction effect, the security indexes, including the wheel/rail vertical force, lateral wheelset force, deraiment, and wheel unloading, become larger.

0 3 6 9 12-60

-55

-50

-45

-40

15

Co-simulation Off-line simulation

Forc

e (k

N)

Time (s) Fig 8. Side force comparison of head train

0 3 6 9 120

20

40

60

80

100

15

Co-simulation Off-line simulationD

ispla

cem

ent (

mm

)

Time (s) Fig 9. Lateral displacement comparison of head train

Tab 1. Comparison of train security indexes

Method Wheel/rail vertical force

(kN) Lateral wheelset

force (kN) Derailment Wheel unloading

Off-line 94.69 28.79 0.31 0.63 Co-simulation

98.92 32.85 0.35 0.78

This means that the fluid-structure interaction cannot be neglected in the calculation of


train security.

V. CONCLUSION

In this paper, an improved algorithm of high-speed train fluid-structure interaction under crosswind is presented. First, data communication of the fluid solver and structure solver is avoided by inserting the program of the train-track coupling dynamics into the fluid dynamics program; second, the load boundary of the fluid-structure interface is improved by introducing the relaxation factor. For the velocity of crosswind of 13.8 m/s and a running speed of train at 350 km/h, the aerodynamic forces and attitude of the head train are compared via the off-line simulation and the co-simulation. The comparison shows that the fluid-structure interaction has a significant influence on the head aerodynamics and attitude, and the security indexes become larger in the fluid-structure interaction simulation. Thus, the fluid-structure interaction calculation is necessary for high-speed trains under crosswind conditions.

ACKNOWLEDGEMENTS

The research was supported by the National Natural Science Foundations of China (Nos.50821063 and 50823004), 973 Program (No.2007CB714701) and the Fundamental Research Funds for the Central Universities (No.2010XS34).

References [1]. X.B. Li, Z. Yang, J.Y. Zhang, W.H. Zhang, Aerodynamics properties of high-speed train in strong wind, Journal of Traffic and Transportation Engineering, 2009, 9(2): 66-73 (in Chinese). [2]. H.Q. Tian, Study development of train aerodynamics in China, Journal of Traffic and Transportation Engineering, 2006, 6(1): 1-9 (in Chinese). [3]. M. Suzuki, K. Tanemoto, T. Maeda, Aerodynamic characteristics of train/vehicles under cross winds, Journal of Wind Engineering and Industrial Aerodynamics, 2003, 91: 209-218. [4]. C.J. Baker, J. Jones, F. Lopez-calleja, et al, Measurements of the cross wind forces on trains, Journal of Wind Engineering and Industrial Aerodynamics, 2004, 92(7): 547-563. [5]. S.G. Tan, X.B. Li, Z. Yang, J.Y. Zhang, et al, The flow field structure and the aerodynamic performance of high speed trains running on embankment under cross wind, Rolling Stock, 2008, 46(8): 4-8 (in Chinese). [6]. B. Dierichs, M. Sima, A. Orellano, et al. Crosswind stability of a high-speed train on a high embankment, Journal of Rail of Rapid Transit, 2007, 221(2): 205-225. [7]. Y.G. Wang, K. Chen, Effects of crosswinds on curve negotiation of high-speed power cars, Journal of Southwest Jiaotong University, 2005, 40(2): 224-227 (in Chinese). [8]. Y. Song, Z.S. Ren, Research on dynamics performance of high speed trains under strong lateral wind, Rolling Stock, 2006, 44(10): 4-7 (in Chinese). [9]. T. Li, J.Y. Zhang, W.H. Zhang, Performance of vehicle-track coupling dynamics under crosswinds, In: Proceeding of Railway Motor Vehicle Dynamic Simulation, Chengdu, 2010: 317-322 (in Chinese). [10]. J.Z. Yang, H.Q. Bi, W.M. Zhai, Dynamic analysis of train in cross-winds with the arbitrary Lagrangian-Eulerian Method, Journal of the China Railway Society, 2009, 31(2): 120-124 (in Chinese). [11]. T. Cui, W.H. Zhang, Study on safety of train in side wind with changing attitudes, Journal of the China Railway Society, 2010, 32(5): 25-29 (in Chinese). [12]. W.M. Zhai, Two simple fast integration methods for large-scale dynamic problems in engineering, International Journal for Numerical Methods in Engineering, 1996, 39(24): 4199-4214♦


NEW APPROACHES TO THE CHOICE OF ARCHITECTURE FOR A SUPERVISORY SYSTEM OF GAZ TRANSPORTATION

BOGDANOV N.K ZAMYTSKIKHY P.V KHADEEV A.S State Technical University – MADI, Moscow, Russia

Summary: The paper presents the advantages of centralized processing of operational information technology systems in supervisory control of gas transport. We consider the technical aspects of data transmission between the equipment automation of gas mains, central control point (DP) and DP slave branches. The article is published with the financial support of the Government of the Russian Federation (Russian Ministry of Education) as part of the project under the Contract 13.G25.31.0064 on October 22, 2010.

Keywords: Dispatch and transport the gas control system, a hierarchical approach, data transfer protocols.

Beginning in 2010, JSC "AtlanticTransgasSystem" participates in a number of projects to create supervisory systems transport natural gas, which created or redeveloped technical infrastructure automation and communication, as well as articulated by the customers job to cause formation of a new design trend in architecture dispatching systems. The traditional approach to building a control system of gas transportation companies across the region was a hierarchical approach "bottom up" when the control towers (DP) of individual industrial branches (linear production controls gas pipelines-Linepipe Operation Center) was concentrated to a maximum of information technology as the linear part of MG, compressor shops, gas distribution stations, etc., as well as data from all support systems-electrical, electro-chemical protection, communications and security. Further, each of DP higher level of management receiving a subset of the technological information available to the DP slave, plus a set of aggregate parameters and data reporting manual input [1]. This concept has several advantages, including:

1. The creation of enterprise control system could be implemented in stages over several years at the expense of major repairs.

2. Control systems not subject to special requirements for the communication channels-Data collection was carried out with grassroots systems over dedicated copper cables within the prom.

The site; with remote control systems-anti-interference, with the use of protocols [2]. Failure to transfer data to the upper level was not considered as critical as branch manager is still in full control of all process and support systems in their area of responsibility and could


report to the central control room over the phone. However, the construction of systems "bottom up" has its drawbacks, as it is impossible to ignore the fact of providing facilities MG, especially newly constructed, in principle more reliable and more than three orders of high -speed fiber-optic lines (FOCL). It features fiber-optic, as well as the investment nature of most of today's development projects and reconstruction of the supervisory system can be considered an alternative to the traditional layered architecture.

The concept of a centralized architecture

Centralized supervisory control (see fig 1) provides the following direction of information flow.

Fig 1. Information flowscontrol and management

1. Process data from the controllers installed at the sites of the linear part of the MG (DCS and RTU-unit, as well as ACS GDS) go directly to the CDP process control. Thus the data are collected as a process of gas transport, as well as state support systems.

2. Manager of branches observe the state of objects in the MG and its neighboring areas of responsibility, using the actual remote terminal access control system CDP. They can also generate reports for yourself - using a common portal. In Fig. 1 within the systems of automatic control, of course, comes to the control operator of the shop. However, some technological information, which characterizes the condition of the equipment in time, goes to the CDP process control.- Data from the gas metering stations, similar to the compressor-nym shops, arrive simultaneously at a local point of data collection and control system in the CDP.

3. Commands closures Manager submits CDP controlling MG in general. Staff offices shall be notified. Teams that require changes in the operational conditions are fed to the level of branches in the form of traffic control tasks. Achieving the targets set automatically controlled by the CDP.

4. Reporting data on supply, use and loss of gas in the area of responsibility of the branch can be formed automatically in the CDP process control, but must be confirmed by the responsible person (manager) of the branch, which can also, if you have the justification to adjust.

What are the benefits and risks is a central passage? First, the decline in the number of program - technical complex control system. This eliminates the problem of PCS service


branches, which are often on the ground there are no qualified personnel. Disclaimer of data storage and processing at the branch level and eliminates the need to synchronize both content and structure of object databases, SCADA - systems - the basic elements of control systems. And if synchronization valuesin principle not a problem (only time - consuming configuration information Interface), the automatic or even automatic synchronization of database structures is almost never realized, that leaves room for defects associated with the "human factor". Secondly, the centralization of facilities allows for a more professional management of their operation - place in the overall data center (DPC) of the organization and use the same highly reliable means of electricity, air conditioning, etc. It becomes easier to apply uniform security policy, both in the field of access to equipment/data, and reduce the vulnerability of the system - for example, to conduct regular backups. Thirdly, the centralized processing of information makes it easier to apply the same algorithms to model data sets of different branches. Thus, if the planning phase costs of FER in the coming period costing process needs its own gas by a group of centralized payment, the owner of a detailed plan for transport/distribution of gas, and the actual values, for example, made of the commodity transport, or loss of gas can be calculated in accordance with the approved method of price-decentralized.

Access Managers of branches to a centralized system will allow them to also see the results of the simulation MG, often carried out for all the gas company as a whole and has a much smaller sense for the plot area of responsibility of a separate branch. Direct access to the DCS/ RTU - CDP devices of the control system also allows you to implement software control of MG and leak detection.

Centralize the management of risks associated with possible failures of hardware and software, and communication channels are rarely evaluated with a formal precision. Reduction in computer technology allows almost every project, set the task of creating redundant systems, with 100% redundancy of servers, workstations, network equipment and cable lines within buildings (areas). For dispatch systems across the region and also back above control towers, geographically remote from the main. This, as well as duplication of data channels through the use of satellite channels, let’s talk about disaster recovery systems established, large-scale failure are possible, but again only because of the "human factor".

Another objection to the centralization of information is 100% of the impossibility of one man-Manager CDP effectively taken a wealth of information and adequately reacts to it. To ensure basic ergonomics PCS CDP - screen form, which works manager must contain, of course, only the most important parameters, and some generalized symptoms, on the other hand, when an abnormal situation. Manager will receive all of the possible amount of information on specific object opening the appropriate detailed mimic. The transfer of information we consider separately the two technical aspects of information interactions in a centralized architecture: a unified communication protocol that is ideal for centralized control system, and the mechanism of dispatching jobs. Any person professionally connected with the industrial automation, sign, or at least heard about the technology of communication OPC, have received very widespread in


the world and in Russia over the past 15 years. In principle, the transfer of air traffic control tasks can be viewed as some internal company e-mail, with the following features:

- For each task should be fixed status, in which it is currently (created, approved, the work performed, etc.), and time of each assignment status;

- A job can contain both text-only instructions and a link to a formal setting, a target which is to be achieved (see fig 2).

- One supervisory task can be associated with others.

Fig 2. Targetsettingin the controltask

CONCLUSION

Considered by the example of the "center-branch" approach can be more successfully applied in the case of three or more levels of supervisory control-remote access to a centralized system, engineers can get a replacement unit, operating without the linear sections MY COP, or managers of individual pipelines. The availability of reliable and high-speed communication channels, modern communication technology, supported by specialized software tools supervisory control, open up the possibility of increasing the reliability and convenience of the new systems under tight control of their budgets as at the implementation stage, and subsequent maintenance.

References

[1]. Berner, LI, Bogdanov, NK, Kovalev AA, Integrated automation solutions for gas transportation and gas companies of OAO "Gazprom" / / Gas Industry. 7, 2007. - C. 38-43.Two. [2]. Zeldin, M., Kovalev AA, Implementation of joint information in the automation of dispatching management (for example, projects for OAO "Gazprom") / / Industr ial process control and controllers. 6, 2008.Three.

[3]. BogdanovA.N., KiselevA.OPC, Unified Architecture: changes in the popular technology information exchange with the engineer's point of view / / Modern automation technology. 3, 2010. - C. 82-87♦


IMPROVEMENT QUALITY OF CONGESTION CONTROLLER IN ATM NETWORK BY METHOD USING NEURAL NETWORK

TRAN XUAN TRUONG

Telecommunication Engineering Department, UTC

Summary: This paper presents method of congestion control using neurons at the switching nodes in the ATM network to improve quality of controlled system. Neural network have a important special role in the prediction the congestion status for the controller, which based on that it can determine the limited rate at the moment of the switches. This is foundation to control the growing traffic source on the network. The experimental simulations have demonstrated the effectiveness of this method when compared with conventional methods.

I. INTRODUCTION

As you known, the ATM network is a network based on cell switching technology (small and fixed length packets, 53 byte) [2]. This is the information network that can provide multiple services, speed from a few hundred Kbps to Mbps. On the other hand, the ATM Forum has five current services for ATM networks but only available bit rate ABR and UBR services is designed to manage the application busrt type data. ABR service is trusted to manage better the data flow, because it allows the switching monitoring network behavior and feedback of relevant information by RM cells to the source. The source will adjust its data rate based on the received information. The task of this traffic management is necessary, because this is the essence problem for network control and distribution efficiency, negotiation about quality of service QoS. Congestion control is one of the most important task in the traffic management. It is a dynamic problem that can not be resolved by conventional static solutions. Typically, only bit explicit forward congestion indication EFCI is used for congestion control purposes. In addition, studies have shown that explicit rate ER can provide better and faster quality, the researchers gave different directions in computing the explicit rate ER as NIST ER [4] and EPRCA [6] techniques. However, these techniques also present limitations when applied in the switching nodes. This is reason we offer solutions to solve network congestions effectively by neurons.

This paper presented a congestion control solution for available bit rate services (ABR) in the ATM network using Neural network. This is a congestion control method indicated explicit rate (ER) by Neural network, which uses Artificial Neural Network to determine the explicit rate for the ABR service. ER is defined as maximum allowed cell rate for source in the next cycle. Neural networks used in this technique will monitor the buffer queue status of the switch and make the suitable ER value carried by resource management cells (RM) to source for adjustment their speed. We perform comparing the proposed algorithm and some known algorithms based on process system modeling and simulation. Since we can see the advantages and disadvantages of this intelligent control method.


II. CONGESTION CONTROL SOLUTION BY METHOD OF THE NEURAL INDICATION EXPLICIT RATE IN ATM NETWORK

The proposed switching congestion control is explicit rate technique indicated by neurons. The main purpose in this design is made Neural Network to create congestion control process simply but effectively. The ordinary congestion control studies can not meet control quality in ATM networks. Several other researchers have proposed the use of Neural Network [5], [7], but they focused on the using Neural networks to forecast traffic patterns. So this section, the author gives a complete algorithm used Neural network to monitor network status and determine how much bandwidth can be seized for.

As mentioned, this technique comply with the specification by the ATM forum, so basic operation is similar to the other switching diagram (see fig. 1). The main difference is way that how the explicit rate values measuring made are. Neural Network will monitor the queue status in the switching buffer for each fixed time period (N). After N received ABR cells, Neural Network should value suitable ER based on the reading information about the queue length for the current and the next time period. When a backward resource management cell (BRM) go to the switch, the calculated values are compared with ER values in BRM cells. If the calculated value is smaller than the current value of ER, the ER field in BRM will be updated. Conversely, if the calculated value is greater than the current value of ER, the ER field in BRM cell will be guaranteed.

Our algorithm has two inputs required to bring the neural network. They are the current and pass queue length. Two input are essential in providing sufficient information about the status of the neural network. Because, from the information about current queue status, Neural network can learn queue occupancy and the level of congestion. Besides, Neural network compares the current queue length with the previous queue length to get information about the buffer change rate. This is a good measured value to known the network congestion ability. If the network is not congestion, this is a good indicator for the network to achieve the completely application of network resources.

Fig.1. The model of the proposed Neural network control technique

ANN

Switch ATM

Queue

Switch ATM

Queue

Neural network

congestion control

Neural network

congestion control

Source DentinationFRM FRM FRM

BRM BRM BRM

ANN


The Neural network output is the rate adjustment factor (RCF) which can be used by the virtual circuit (VC) based on current traffic conditions. The value of this coefficient ranges from -1 to 1; In there, the value is less than 0 displays the network facing congestion or have the potential to be network congestion, so the source has to transmit at rate lower than the current cell rate. Otherwise the RCF value is greater than 0 displays all links that can be used completely, and the speed of the source can be increased to match the traffic of the suffer network. With the RCF provided by the network neurons, explicit rate ER for a new cycle is calculated by:

ERnext = Min(Link Rate, max(0, ERcurrent + RCF x Link Rate)) (1)

All switches will ensure an value of ER and updated in each N received ABR cells. For each cell passing through switches, the switch examines the ER field of BRM cells. If the ER field value of BRM cell is greater than explicit value that switches ensure, the ER field in BRM cell is updated to ER ensured value of the switch. If the value in the BRM cell is smaller than do not perform any action.

For behaviour at the source, they will adjust its data rate based on the value of this ER with the help of congestion indication bit (CI) and no increase indication bit (NI) in the same cell. The allowed cell rate of the source in each control cycle is defined based on the bit values of NI and CI in the resource management cell BRM. When the NI bit is not set (NI = 0), the are allowed cell rate at the source is defined in value:

(2) ⎢⎣

⎡===+=

1CI if ACR) x RDF - ACR min(ER, ACR0CI if PCR) PCR, x RIF ACR min(ER, ACR

When NI was established (NI = 1), the allowed cell rate at the source is determined by:

(3) ⎢⎣

⎡====

1 CI if ACR) x RDF - ACR min(ER, ACR0 CI if ACR) min(ER, ACR

III. CONSTRUCTION NEURAL NETWORK USING IN THE PROPOSED CONGESTION CONTROLLER

As pointed out in Section 2, the neural network is a very important role in predicting the network congestion level based on the measure the switch queue value between two consecutive time. On that basis, at a time, Neural network will provide a ER value to switches through the RCF from the neural network output. Therefore, the development of Neural network to undertake its role in the congestion control is essential.


y(k)

1

a1(k) = logsig(IW1,1. p1(k) + IW1,2. p2(k) + b1) a2(k) = logsig(LW2,1. a1(k) + b2)

p2(k)

p1(k)

Class 1 Class 2

a2(k)

12x1

a2(k) a1(k) n2(k)

Class 3

n3(k)

5x1

12x1

a3(k)

6 12

5 5x1

5x1

12x1 6x1

1 12x6

12x1 1

n1(k) 6x1

6x1

6x1

IW1,1

IW1,2

b1

LW2,1

b2

1 5x12

LW3,2

b3

a3(k) = tansig(LW3,2. a2(k) +b3)

6x1

Inputs

n4(k)

1x1

1 1x1

Output 1x5

LW4,3

b4

y(k) = purelin(LW4,3. a3(k) +b4)

Class 4

Fig. 2. The structure of neural network used in the study

Based on the Neural network knowledge, the design skills to build the network neurons, we

have built a network neurons completely to apply in the proposed congestion control problem.

Neural network are designed that has two inputs for receiving the corresponding value of the

queue length at the previous and the present time, separated by a control cycle. The distance

between that time was the distance of two cell RM. The obtained neural network output

parameter is ratio RCF to help switches decide it’s explicit rate ER at next time.

Neural network architecture consists of four layers and described in Fig.2, in which the

input layer (layer 1) has 6 neurons; Layer 2 has 12 neurons; layer 3 consists of 5 neurons and

the output class (class 4) has 1 neuron. The neurons in the first and second layers used the

transfer function as logsig, neurons in layer 3 used the transfer function as tansig, the remaining

neuron in the output layer used transfer function as purelin form. The set of weight matrix

including a threshold, class 2 and class 3 weight matrix (net.b 1, 2 net.b, net.b 3, 4

net.b), input weight of the two inputs to the network’s first layer (net.IW 1.1; net.IW 1, 2),

connections weight between layers (net.LW 2.1, net . IW 3.2, 4.3 net.LW) are given

detail in [3].


IV. BUILDING THE SIMULATION MODEL TO VERIFY THE PROPOSED METHODOLOGY

Fig. 3. Configure an ATM network using simulated and compared

Link 155Mbps

SCBR

DABR-C

DCBR

DABR-D

DABR-A

DABR-B

SABR-A

SABR-B

SABR-C

SABR-D

S1 S2

With the Neural network has been designed in Section 3 and the algorithm presented in Section 2 above, we have used broadband communication network ATM as shown in Fig.3 to perform this simulation. The network consists of four virtual channels (VC) bring available bit rate service (ABR), called group A, B, C and D. Besides, it contains four VC carrying the service constant bit rate (CBR) named group X. The network has two switching nodes and are connected by 155 Mbps links. The ABR source is set at the rate of 31 Mbps, 93 Mbps, 155 Mbps, 217 Mbps and 279Mbps respectively while the CBR source is fixed at the rate of 31 Mbps for all simulated cases. The CBR source has main function to occupy a bandwidth certain amount in the link and also to observe its impact to the network configuration. For switches, they have time slot of 155 Mbps with buffer size of 900 cells. There are two reasons for choosing the small buffer size. First of all, this is the design trend to keep the switching buffer size to be smallest. In addition, the smaller buffer size can make us observed congestion control scheme when the source has a minimum rate entirely possible.

The link between broadband terminal equipment (BTE) and switches is set to 2 km with rate of 155 Mbps. The links between the switches is also 155 Mbps but the distance between them is 10 km. Simulation time for each run is 500 ms, while the received cells per receiver and the lost cell number in each switch are recorded. In this simulation, only the measure about quality of service (QoS) is observed, which is characteristic of the control rate range and queue in the switch node with the neural network controller. Since we can calculate the throughput of the switches.

Besides the proposed explicit rate neural network congestion control algorithm, the simulation is repeated using the algorithm that is accepted by the ATM Forum, which is a Enhanced Proportional Rate Control Algorithm (EPRCA) [6].

V. THE OBTAINED RESULTS FROM SIMULATION AND DISCUSSION To obtain the necessary results from the simulation system, we undertook to write the

simulation program based on Matlab software version 7.8 in 2009. The traffic type can be selected as the source with Poisson distribution or two-state Markov distribution, the other


hypotheses as shown in section 4. To conduct a survey of control rate range for resources, we have obtained the results

presented in figures fig.4, fig.5 and fig.6. The results showed that control method by Neural network has source control rate range wider and larger than the conventional control methods. The rate range of proposed method has values from 5Mbps up to 220 Mbps, whereas the conventional method only from 10 Mbps to 200 Mbps. This parameter also reflects partly the high sensitivity of the control method using Neural network. This result due to the congestion predictability quite well of the neural network in controller, the more accurate forecast, the more accurate value of ER in the switches, which the ABR sources the can adjust their rate to the network accurately based on these reference rate.

0 50 100 150 200 250 300 350 400 450 5000

255075

100125150175200225250

Time [ms]

Source Rate [Mbps]

Fig. 4. A range of control rate in method using Neural network

0 50 100 150 200 250 300 350 400 450 5000

255075

100125150175200225250

Time [ms]

Source Rate [Mbps]

Fig. 5. A range of control rate in the conventional method

0 50 100 150 200 250 300 350 400 450 5000

25

50

75

100

125

150

175

200

225

250

Time [ms]

Source

Rate [M

bps]

Neural Control MethodConventional Method

Fig. 6. Compare a range of control rate in two methods

The queue properties of the switching nodes using neural control technique as fig.7 shows that the buffer level is often consumed at high levels but concentrated in the safety of queue overflow problem and the queue’s cycle characteristics. This feature demonstrates, the ability using network resources to be effective and therefore the switches throughput will increase significantly when compared with other control methods. At the same time it can avoid the cell loss phenomenon due to full buffer queue, so it will contribute to reducing the rate of cell loss compared to many other control methods. Meanwhile, for the conventional control method, the variation of queue length always permanents in a dangerous level, in which the switch will overflow buffer and the throughput ability in the buffer also be slowly. This decreases the switch throughput and be easy to cause the cell loss due to the full queue, which affects directly the quality of service of the applications.


0 50 100 150 200 250 300 350 400 450 5000

100

200

300

400

500

600

700

800

900

Time [ms]

Queue Length [cells]

Neural Control MethodConventional Method

Fig. 7. Characteristics of queue in the switch under different control techniques

0 31 93 155 217 279200

220

240

260

280

300

320

Source Rate [Mpbs]

Throughput [Mpbs]

Neural Control Method

Conventional Method

Fig. 8. Throughput of switches in the control methods

For the throughput of the switches, we can observe the results obtained from the fig.8. The switching throughput of the methods changed according to the rate, the Neural network control method achieves high throughput (300 Mbps), meanwhile, conventional control scheme as EPRCA has lower level (about 275 Mbps). This property also shows the traffic congestion control ability be better with conventional methods, because the throughput is the total traffic to be served by the switching node so the traffic control as efficiently the traffic through the switches as many, also means higher throughput. Furthermore this will avoid the network collapse phenomenon (a throughput be reduced in minimum, negligible value) due to the traffic retransmitted too much from the sources because the sender - receiver process be failed.

VI. CONCLUSION

In this paper, we have implemented the congestion control method using Neural network applied for the ATM network. It is demonstrated that the proposed control algorithm is better when compared to the scheme switching using conventional control methods, because it has the quality of service better. It also has the simplicity and efficiency, because the simulations have shown that the proposed algorithm has high network throughput, the high link utilization and low cell loss rate. To have an expected result because the neural network can predict accurately about congestion in the high volatility and unpredictable traffic environment of ATM network.

References [1]. Bùi Công Cường, Nguyễn Doãn Phước, Hệ mờ mạng nơ ron & ứng dụng, Nhà xuất bản Khoa học kỹ thuật, 2001. [2]. Nguyễn Hữu Thanh, Tổng quan về kỹ thuật mạng B-ISDN, Nhà xuất bản Khoa học Kỹ thuật, 2007. [3]. Trần Xuân Trường, “Nghiên cứu giải pháp chống nghẽn trong mạng thông tin băng rộng ATM sử dụng mạng nơ ron”, Đề tài nghiên cứu khoa học cấp Bộ, nghiệm thu năm 2010. [4]. Golmie, N., Chang, Y và Su. D, “ NIST ER Switch Mechanism”, ATM Forum/95-695. [5]. N. J. H. Kotze and C. K. Pauw (1997), “The use of Neural Networks in ATM”, Proceeding of the Communications and Signal Processing, COMSIG '97, South African, Page(s): 115-119. [6]. L. Robert (1994), Enhanced PRCA (Proportional Rate Control Algorithm), ATM forum 94-0735 R1, August 1994. [7]. A. Tarraf, I. HabiB and T. Saadawi (1995), “Congestion Control Mechanism for ATM Networks Using Neural Networks”, IEEE International Conference on Communications, ICC '95 Seattle, Volume: 1 Page(s): 206-210♦


ENVIRONMENTAL INDICATORS FOR SUSTANABLE DEVELOPMENT URBAN TRANSPORT PLANNING IN VIETNAM

MSC. VU KIM HUNG Institute of Transport Planning and Management University of Transport and Communications

Summary: The research aims at selecting environmental indicators for the planning of sustainable development of urban transport in Vietnam. The author used the methodology of systematic analysis, methods of literature review on environmental indicators in the development of urban transport in the world and DPSIR model to assess the necesity to consider the environmental indicators in the process of urban transport planning. The article also described the selection of the 13 environmental indicator based on 4 criteria for protection of natural resources and environment.

Key words: Environmental indicator; environmental criteria, urban transport planning.

I. INTRODUCTIONS In general, big cities in Vietnam are facing with challenges in developing urban transport.

Problems of urban transport has become particularly severe, in particular, traffic jams are becoming more frequent and extended over almost the city. Next is the phenomenon of flooding due to poor sewerage during the heavy and long-lasting rains. On the other hand, the increase means the number of private property led to the demand for fossil fuel. Increasing greenhouse gases and air environment, dust, noise, vibration due to operation of urban transport is becoming increasingly serious, threatened ecosystems, urban green space narrowing leads to loss of ecological balance.

In fact, urban planning and urban transport planning have been developped, but there is no standard for developping the plan. Almost planning of urban transport development are lack of sustainability, lack of strategic vision and un-friendly with the environment.

Law of Environment in 2005 has already required to assess the environmental impact strategies for strategic projects, master plans and planning of transportation. However, this work still has a lot of inadequacy in the coordination between the transport planning and environmental impact assessment strategies.

The above-mentioned problems show the necessity to incorporate environmental indicators into the development process of urban transport which must be taken into consideration from the planning stage. By the law of Vietnam shall all urban transport projects be taken into construction only after being approved in the steps for urban transportation planning.

The objective of this study is to identify and select indicators of environmental conditions suitable for urban transport development in Vietnam towards the goal of sustainable development.

However, in this study, the author do not quantify the selection indicators and general


assessment of interaction with the target of economic and social indicators.

In order to see the relationship between environmental indicator with the objective of urban transport development, the environmental indicator for the urban transport planning is defined in the tree of criteria identifying (see figure above).

Definitions Sustainable Development in the report of the

United Nation Commission on Environment and Development (United Nations, 1987) [2] is defined as follows: "Sustainable development means development that meets current needs but without prejudice to the ability to meet the needs of future generations".

Indicators

Criteria

Vision

Objective

Urban Transport is a system including the interaction of such three subjects as transport infrastructure (roads, transit pointa, bus stops/transitions), transportation and traffic control systems in order to the satisfy the mobility demand of human (passenger x km) and freight (ton x km) [1].

Indicators of environmental sustainability in urban transport has been studied in the world such as planning policy for urban transport development of the European Union [3]; In the report of the Ministry of Environment and Energy, USA about “Indicators and performance measures for Transportation, Environment and Sustainability in North America”[4]; In the conference of OECD about Environmentally Sustanable Trasport (EST) in Vienna “Futures, Strategies and Best practice” [5]; And in a Guidebook for Performance-Based Transportation Planning of transportation research Board (TRB) [8].

II. METHODOLOGY

This study is carried out based on the following methodology: Author used DPSIR model to assess the status of the urban environment in Vietnam. The

author selected urban transport in Hanoi as the object of study and assessment, suggesting that environmental indicator should be considered during the planning process of urban transport development.

DPSIR model inherited by the European Environment Agency (EEA) is one of the basic frame, chain facilities for general information, using the indicator with the different categories (driving force - Pressure -state of Environment - Impact - Response) (UNEP / RIVM, 1994) [6]; (RIVM / UNEP,1995) [7]. This model is similar to the model PSR framework (OECD, 1993 Organization for Economic Cooperation Development) but with two (D) item is the driving force and effect (I). DPSIR is the first letter of the year from the English language:

Driving Forces (D) is the general motivation factors are affecting the local environment are being considered. For example, sustainable urban transport development.


Pressure (P): The pressure of dynamic factors on the environment and alter the current status of environmental pollution. For example: Increased emissions, dust, noise, vibration, increased greenhouse gas,…

State (S): The state of the environment at a time or time period. For example, the current stateof air (SO2, NOx, TSP, VOC), noise, vibration, loss of biodiversity ....

D

P

S

IR

Impact (I): The impact of negative or positive to the economic situation, social and human health. Example: Due to increased pollution of traffic leading to increased incidence of disease, ...

Response (R): integrated solutions to improve environmental quality. From the study of environmental indicator in the world and the lack of environmental

indicators in transport development planning of Hanoi. Author identified and selected appropriate environmental indicators for the planning process of urban transport development in Vietnam.

III. RESULTS OF RESEARCH

Through the study authors go to some of the results as follows: Environment indicator is understood as a functional description of environmental issues

and resource development planning in urban traffic, the data base to be able to conclude the planning impacts the quality the environment.

Sustainable urban transport environment means developing an urban transport system to meet the transportation needs of the market and people travel now and in the future, saving land use restrictions energy consumption and reduce emissions and waste within the limits of absorption of the medium, with the mode of travelaffordable alternative to reduce traffic congestion.

It is easy to see the lack of system of environmental indicators in urban transport development planning processes. The author used the DPSIR model for analysis and evaluation. Driving force factors selected by the author is to develop sustainable urban transport and environment is expressed through the secondary driving force is to develop infrastructure, vehicle development and traffic control. From The Driving force of urban transport growth above will cause Pressure on the environment such as the land use, energy use and emissions in transport, traffic congestion…. leading to State environmental quality, expressed through changes of parameters such as SO2, NOx, VOC, TSP, PM10, Pb and noise, greenhouse gas and landscape. The Impact of environmental pollution are analyzed through the economic damage and social impact on public health. With the impact that requires, Response is the integrated solution to improve the environmental quality of urban transport cause, that is solutions and policies to achieve development objectives of urban traffic and minimize impacts environment, is shown in the image below.


Through the DPSIR model shows that environmantal indicators are important toward the

urban transport system for sustainable development. Starting from the perspective (vision) for the development of each country, urban or community that is sustainable development, defined in the Brundtland report in 1987 at the United Nations as: "development to meet present needs without compromising the ability to meet the needs of future generations".

Urban transport development - Infrastructure development - Vehicle development - Traffic control

Driving Force (D)

- Land use - Energy - Emission and congestion

Pressure (P)

- Transport emission - Greenhouse gases - Landscape

State (S) - Land use for transport policy - Air Quality Management - Energy quality management - Transport demand management

Response (R)

- Public health - Economy damage

Impact (I)

To generalize the sustainable development of urban transport can be understood as: “urban transport development to meet current trafic demands without compromising the

ability to meet the transport needs of future generations”. Regarding the demand for urban transport in the sustainable development of urban

areas there are many ways define different but come from a "model for the metabolism of the settlements of man (human stetllement's metabolism model), Newman and Kenworthy (1999) suggested that the path reaching a state of sustainable development, each city needs to achieve three objectives: 1- there is adeveloping economy, efficiency and environment friendly, 2 - have a rich culture, harmony and charming, and 3 - have a transportation system for sustainable urban transport [1].

Inherit the results of different studies, the authors [1] has compiled an objective framework for sustainable development of urban transport system consists of 4 levels. The highest level is the vision of a sustainable urban development, such as the Brundtland concept, followed by the objectives of sustainable urban development, as defined by Newman and Kenworthy (1999). Three is the level of the concept of an urban transport system's sustainability and members Albert Speer (1993) is a transportation system to achieve the following objectives: 1 - Traffic smooth, 2-traffic safety, 3 - environmental transportation friendly, 4 - transport to promote economic development. Number 4 levels are 13 criterias to be achieved to have a transportation system for sustainable development including 4 criteria for traffic movement; 2 criteria ensure traffic safety; 4 criteria for protecting natural resources and emvironment and 3 criteria for improving economy of the city and region.

Based on the target framework above, in terms of the environment in urban transport shall be four criteria on environment and natural resources of concern are: 1-


reduction air emissions criteria; 2- reduction fuel and energy use in transport ; 3-criteria for land use for urban transport and 4- criteria to save space and landscape. Based on the criterias above, the authors selected 13 environmental indicator for the sustainable development of urban transportin as shown in the picture below:

Level 4: Criteria for Sustanable Urban

transport Development

Sustanable Urban Development

Sustanable Environment

Economically Viable&

Efficient

Socially compatible

Leve l 1: Vision

To achieve a strong and compact

Economy within urban area

To achieve a Sustanable Urban

Transport System

To achieve a rich, harmonise and compact culture within urban area

Level 2: Objective for

Sustanable Urban Development

To ensure mobility for all

transport demand

To ensure safety of all transport movements

To protect natural resources

and Environment

To improve economy of the

city and region

Level 3: Objective for

Sustanable Urban transport

Development

To reduce air pollutions

To reduce traffic noise

To save energy consumption in transport

To save space consumption for transport

To ensure equality in using transport properties

To increase the number of

modal choises

To increase productivity and efficiency of transport

supply

To reduce accident

frequency

To reduce accident

severity

To reduce tatal transport cost

To increase economic productivity and

efficiency

To improve economic

attractiveness

To increase capacity of transport supply

Level 5: Indicator for

Sustanable Urban transport

Development

(1)-Emission per vehicle (ton/vehicle) (2)-Emission per each route (ton/km) (3)-Tons of greenhouse gases generated (4)-The level of noise (dB/km) (5)-Fine dust (ton PM10/km)

(6)-Fuel consumption per VMT (km/vehicle) (7)- Fuel consumption per PMT (km/passage) (8)- Fuel consumption per freight (km/ton goods)

(11)-Percent of change in Public transportation per other mode (km/km). (12)-Percent of vehicles using alternative fules (%). (13)-Percent use of non-motorized modes per all trips (%)

(9)-Percent of urban land change in urban transport (%) (10)- percent of trees in street (%)

Ghi chú: [1]; Author's choice

1. Environmental criteria related to traffic emissions: The objective of this criterion is to reduce emissions on vehicles and on roads. To achieve this criterion we need to control and quantify the emissions of each of the planning compared with the current state plan and is expressed through the indicator of (1) (2); (3) (4) (5).


2. Environmental criteria related to reduced fuel use in transport and energy: The objective of this criterion is rational use of energy in the transport of passengers and freight. Thus, for transport planning of urban, we need to identify targets for the rate of fuel consumption per unit of transport is made or tons of kilometers of each mode of transport urban centers. Only when the energy cost per unit of urban transport is reasonable pollution reduction can be expressed through the indicator of (6) (7) (8).

3. Environmental criteria related to rational use of urban land and environmental landscape, is reflected by the indicator of (9) (10).

4. Environmental criteria related to alternative modes of travel: The objective of this criterion reduced the number of personal motor vehicles, increase the percentage of public transport modes and non-motorized modes that will make reduce emissions from transportation activities as well as reduce energy demand in the transport of passengers and freight, expressed through the indicator of (11), (12) and (13).

Above are 13 environmental indicator selected by authors towards an urban transport system of environmental sustainability. These indicators can be reviewed and evaluated directly or indirectly in the process of urban transport planning.

IV. CONCLUSION

The author has provided with the definition of sustainable urban transport in environment. By DPSIR model approach, the author has identified environmental problems caused by transport development in urban in big cities of Vietnam, indicating that regardless of the environmental indicator from step urban transport planning . From general research in the world, author have chosen 13 environmental indicators based on 4 general criteria for sustainable development and environmental resources.

References [1]. Khuat Viet Hung (2006), Traffic Management in Motorcycle Dependent Cities, Doctoral Dissertation, Darmstadt University of Technology. [2]. United Nations (1987), Report of world commission on environment and development, Development and International Economic Co-operation, Brundtland. [3]. Commission of the European Communities (CEC) (1995), The Citizens Network. European Commission Green Paper (COM 95), Brussels. [4]. German Marshall Fund Fellowship - GMF (2000), Indicators and performance measures for Transportation, Environment and Sustainability in North America, Ministry of Environment and Energy, USA. [5]. OECD (2000), Futures, Strategies and Best practice. OECD Conference Environmentally Sustanable Trasport (EST), Vienna. [6]. RIVM/UNEP (1995), Scanning the global environment: A framework and methodology for UNEP's reporting functions. UNEP Environment Assessment Technical Report 95-01, Nairobi, Kenya. [7]. UNEP/RIVM (1994), An Overview of Environmental Indicators: State of the art and perspectives. UNEP Environment Assessment Technical Report 94-01, Nairobi, Kenya. [8]. TRB (2000), A Guidebook for Performance-Based Transportation Planning. National Academy Press Washington, D.C. USA♦


ALGORITHM OF VIRTUAL TRAINING COMPLEX DESIGNING FOR PERSONNEL RETRAINING ON PETROCHEMICAL

ENTERPRISE

BARINOV K. A KRASNYANSKIY M. N MALAMUT A. J OSTROUKH A. V State Technical University – MADI, Moscow, Russia

Summary: The algorithm of design of a virtual simulator that is not tied to a specific discipline is developed. We propose a sequence of development stages of a virtual simulator in accordance with the proposed algorithm, in order to increase productivity and ensure high quality simulator created.

This work was supported by the Government of the Russian Federation (Russian Ministry of Education) as part of the project under the Contract 13.G25.31.0064 on October 22, 2010.

Keywords: Virtual simulator, algorithm, training.

In the study of most subjects it is necessary to support indissoluble connection between theoretical analysis and experimental research. In the traditional method of training experimental research can be conducted in normal laboratory conditions [2]. Current rate of development of distance education has put before the developers the problem of unification of virtual simulators (VS) creation methods. Solution to the problem is possible with common algorithm for the creation of computer simulator that is not focused on any particular discipline.

Modern computer technologies allow automating the process of theoretical knowledge gaining, as it has no significant difficulties in implementing the corresponding software. More difficult in terms of automation is the process of practical skills development [2], as well as learning new methodologies used in the study.

Virtual simulator (VS) designing

For the practical skills gaining virtual simulators can be applied. Simulators can be created with the help of specialized computer systems (constructors) or by a combination of various software tools. Currently, constructors of computer simulators, as independent software packages, allow you to create simulators almost of any subject from scratch, are not widespread. Either way, regardless of a method of simulator creation to be used, it is necessary to add here some algorithm of virtual simulator designing to automate the process and try to consider maximum number of components and featuresin it [3], [4].

In a number of works attempts to develop and formalize the algorithm of virtual simulator


designinghave been made, but they were focused only on a small number of industries: nuclear, petroleum, railway and electrical engineering. Also, besides this disadvantage - arrow focus, it should be noted that there was no multi ser mode provided in a considered number of works. Among other restrictions,an orientation of virtual simulators on operators retraining can be mentioned. In this work we propose extended, yet universal (not tied to any particular discipline) scheme of the algorithm, which eliminates the above disadvantages (partially). The scheme of the algorithm of designing of a virtual simulator for operators of chemical and technological systemswas taken as the base [1].

The process of computer simulator designing, which is an intelligent system itself, requires several tasks to be solved for providing following characteristics of a virtual simulator:

• High degree of similarity between model and real object;

• Possibility of working out the specified actions;

• Ensuring high efficiency for short duration training courses;

• To support for multi-user mode;

• To meet common psychological and pedagogical requirements;

• Short time of adaptation during the transition of human activities from the simulator to the real object;

• Opportunity to make updates and changes without additional costs.

The algorithm of a virtual simulator designing can be represented as a functional diagram in IDEF0 notation.

Sequence of development stages of VS

The inverse decomposition allows us to represent considered algorithm stepwise.

Fig 2. Development stages of VS


Fig 1. Functional diagram of the algorithm of a virtual simulator designing

The first stageis analysis of the subject area. The main steps are:

• Reproduction of complete process flow chart for simulator being created (all that involved in the experiment is to be modeled according to its characteristics; it is a fundamentally important moment, since a completely modeled process allows correctly and fully to work out the required tasks and gain skills);

• Forming of suggestions concerning information technologies and methods of simulator implementation to be used;

• Discussion of acommon interface, design and style of simulator.

The second stage is modeling of structure of VS. The main steps are:

• Development of mathematical models for all devices and processes in accordance with the technological scheme (the creationof a mathematical model should start from its creation for the ideal process);

• Refinement and accordanceof actions to perform work for each component (module)


of simulator;

• Programming of application logic;

• Bringing the training materials, needed for a particular experiment, in accordance with the requirements,developed on the first stage, to its formalized description.

The third stage is setting of VS. The main steps are:

• Introduction the required perturbing factors into the mathematical model, if necessary, to adoptthe model to the real process, considering differential acceleration of the particular processes, as long as for the observation and execution of several processes, depending on the discipline, considerable time resource is required (precisely adjusted mathematical model increases the efficiency of training due to its identity to the real process);

• Designing ofthe tools for training control and management.

The fourth stage is packaging and testing of VS. The main steps are:

• Stepwise treatment of formalized material;

• Connection of the modules in a single complex;

• Testing and validation of execution logic;

• Adjustmentof the simulator for its future exploitation.

Conclusion

Simulator designing with observance of division of processes and stages makes it possible to significantly reduce development time, to improve the quality and reliability, to simplify the process of maintenance, updating and support of virtual simulators. Especially whena virtual simulator is created by a team, such approach gives the opportunity to adapt already developed technology (algorithm) for creating simulators on other disciplines.

References

[1]. V.A. Nemtinov, S.V. Karpushkin, V.G. Mokrozub [i dr.]. Metody i algoritmysozdaniyavirtual'nyh-modeleihimiko-tehnologicheskihsistem: monografiya/M-voobr. inauki RF, GOUVPO «Tamb. gos. tehn. un-t». Tambov: Izdatel'skiidom TGU im. G.R. Derjavina, 2011. 282 s. [2]. Dmitriev V.M., Gandja T.V.. Zadachipostroeniya i konfiguraciyakomp'yuternyhtrenajerov // Distancionnoeobrazovanie, innovacii i konkurentosposobnost': Materialyregional'noinauchno-metodicheskoikonferencii. –Tomsk: TGU sistemupravleniya i radioelektroniki, 2004 – S. 85-86.

[3]. S.K. Gupta, D.K. Anand, J. Brough, M. Schwartz, and R. Kavetsky. Training in Virtual Environments. A Safe,сost-effective, and engaging approach to training. University of Maryland. 2008.

[4]. Robert J. Seidel, Paul R. Chatelier, Virtual Reality, Training's Future? Perspectives on Virtual Reality and Related Emerging Technologies. New York. 1997. 232 s♦


NI DOPING EFFECTS YBA

I. INTRODUCTION

Compound YBa2Fe3O8+w (YBFO) is an antiferromagnet (AFM) with Neel temperatures (TN) around 660 K [1]-[2]. The crystal structure of YBFO is triple perovskite as it is formed by merging Y and Ba, and by disordering of the oxygen atoms and vacancies. The oxygen stoichiometry 8+w varies in a narrow range of −0.2＜w＜+0.1 [3]. The crystal is the only analogue of YBa2Cu3O7 (YBCO) where Cu is fully replaced by another transition metal. There is a wide miscibility gap in two phases as iron is unable to adopt a square-planar coordination. Fe3+ is stabilized in the environment of one octahedral and two square-pyramidal per unit cell, where YBFO composition is generated. YBFO and YBCO differ in their arrangements and the number of oxygen atoms on the basal plane of their unit cell. When w>0, chemical analysis conducted to locate the oxygen in excess of the eight atoms per formula unit showed that the sites available for such oxygen were those in the Y layer at 0, 0, 1/2, assigned as O(4) [4]. The structure diversification of introducing Fe into YBCO has been widely studied in an attempt to detect possible correlations between structural modifications due to the presence of iron and the lack of superconductivity.

Ref [4] discussed the nucleus and magnetic structure of YBFO, samples of which were prepared from liquid-mixed citrate precursors. The cooperative magnetic structure for all compositions was based on a larger cell related to the nuclear cell by the transformation matrix [1,−1,0/1,1,0/0,0,2] [4]. The tetragonal crystal structure of YBa2Fe3O8 became orthorhombic once oxygen vacancies were included into the iron coordination octahedron [4]. The antiferromagnetic ordering of the iron magnetic moments was below 630–670 K, indicating that the variation in oxygen content was too small to significantly affect the magnetic exchange interactions.

2FE O3 8+W

XIAOYU GUAN

YONG ZHAO

XIAOQIU JIA


Summary: By doping Ni into YBa2Fe3O8+w (YBFO) system, we obtained the phase YBa2Fe3-xNixO8+w (YBFNO, x=0, 0.05, 0.10, 0.15, 0.30, 0.50, 1.00). This paper discusses the changes in crystal structural, resistivity and magnetoresistivity (MR) of YBFO samples due to the incorporation of transition metal Ni. The results show that Ni substitution for partial Fe in YBFO does not substantially transform the structure of parent phase, but results in tiny changes in the lattice parameters. The YBFO crystal with Ni doped is semiconducting.

Key words: YBa2Fe3O8+w; Ni doping; crystal structure; magnetoresistivity.


Huang et al. [5] investigated the accommodation of Co in the oxygen-saturated solid-solution phase YBa (Fe Co2 1-z z) O8+w3 using techniques of powder X-ray and neutron diffraction, and 57Fe Mossbauer spectroscopy. In the nominal composition range 0.00≤z≤1.00, the solid-solution limit under syntheses 950 in 1 bar O2 was z=0.47(5) [5]. No symmetric change in the nuclear and magnetic structures was observed due to the Co substitution. The Co atoms were distributed evenly over the two sites, which were square-pyramidally and octahedrally coordinated for w=0. The value of w was viewed as nearly constant and Co adopted valence close to 3.00.

At present, the research on doping transition metals into Fe position in YBFO system, such as Ni and Mn, has not been reported. This paper investigates the Ni doping effects on crystal structure, resistivity and magnetoresistivity (MR) of YBFO system.

II. EXPERIMENTAL PROGRAM

Conventional sol-gel method was followed to prepare samples YBa Fe2 3-xNi Ox 8+w (x=0, 0.05, 0.1, 0.15, 0.3, 0.5, 1, for short YBFNO).

Step 1: Weighed high-purity stoichiometric Y2O3, Ba(CH COO) , Ni(CH COO)3 2 3 24H2O and Fe powders.

Step 2: Raw materials were uniformly and slowly dissolved into acetic solution with continuous stirring during a gradually heating process. The heating process continued until the solution became dry.

Step 3: Grinded the xerogel for 60 min to gain the well proportioned powder and then pressed the powder into tablets.

Step 4: Sintered at 1 100 °C in the atmosphere for 350 h and cooled in the furnace.

To improve the sample homogeneity, intermediate grinding and palletizing must be ensured. The sintering processes are shown in Fig. 1. A Philips X’Pert MRD diffractometer with Cu-Ka radiation recorded the θ-2θ XRD patterns, which characterized the phase purity of the series samples. Each sample was in single phase with tetragonal structure. The resistance-temperature curves were obtained by PPMS (physical property measurement system).

(a) The first fire b) The second and third fire

Fig 1. Heat processes of YBFNO samples

Time (h)

Cool with furnace

Tem

pera

ture

(o C) 1 100 oC

100 h 100 oC/h

500 oC 50 h

Atmosphere 100 oC/h

Room temperature

6 h

1 100 oC

100 h 100 oC/h

Atmosphere

Cool with furnace

Tem

pera

ture

(o C)

Room temperature

Time (h)


Fig 2. XRD patterns of YBFNO

III. RESULTS AND DISCUSSION

3.1. XRD test and analysis

Typical XRD θ-2θ patterns for YBFNO are shown in fig 2. Similarities between spectra of YBa2Fe3-xNixO8+w (x=0.05, 0.1, 0.15, 0.3, 0.5, 1) and YBFO (x=0) reveal that incorporating a small amount of Ni into the YBFO lattice will not substantially transform tetragonal structure of parent phase since only tiny changes of the lattice parameters are observed. In fig 2, the intensity of the (100) peaks gradually become smaller, meaning that the tendency towards a-axis preferential orientation is weakened as Ni content increases. Impurities peaks of Ba2Fe2O5 and BaFe2O4 can occasionally be detected, indicating that they are close to YBFO in phase graph. Only a few impurities peaks of Ni oxides are found. This is because most of Ni is incorporated into octahedral or square-pyramidal environments in YBFO lattice in the substitution for Fe. In addition, in samples of x＞0.3, the typical peaks intensity of objective products are not prominent and the impurity peaks such as Ba2Fe2O5, BaFe2O4, Y2BaNiO5 and NiO become stronger. When x≤0.3, the intensity of characteristic peaks in YBFNO spectra dominates even though there are a few negligible impurities. As a result, we presume that the solid-solution limit of Ni in YBFO is in the vicinity of x=0.3 under the condition of adequate oxygen atmosphere.

Peaks for a series of samples of YBFNO in fig 3 shift slightly to the right so as to conform a strong dependence of peak position on the content and radius of the Ni2+ incorporated into YBFO system.

20 30 40 50 60 70 80

2 (°)

In

tens

ity (c

ount

s)

x=0

x=0.05

x=0.1

x=0.15

x=0.3

x=0.5

x=1 111

Ba2Fe2O5

BaFe2O4

Y2BaNiO5

(1)

(1)

(2) (2) (2)

(1)

(2) NiO

112

005

113

006

020

210

116

213

106

310

109

310

226

100

103

O

O

O


Fig 3. The enlarged diffraction peaks (006) and (200) in YBFNO

For instance, the lattice parameters of YBFNO gradually become smaller as Ni content increases. However, if Ni enters in YBFO crystal and does not substitute Fe, the lattice parameters of YBFNO will be larger than those of the parent phase, and the new crystal will expand. Thus, we are convinced with the partial substitution of Fe by Ni in octahedral and square-pyramidal coordinations of YBFO crystal. The lattice parameters change because the radius of Ni ion is smaller than that of Fe. And the oxygen content in the unit cell is nearly a constant when oxygen is adequate.

Tab 1 lists the values of the lattice parameters versus Ni content within its solid-solution limit. These data are obtained by a least-squares fit to a large number of the Bragg diffraction.

Tab 1. The calculated unit cell parameters of YBFNO

Ni-content Unit cell parameter on the x axis

Unit cell parameter on the z axis Volume of unit cell

0.00 3.916 45 11.813 68 181.20

0.05 3.916 38 11.813 58 181.19

0.10 3.913 02 11.802 57 180.72

0.15 3.910 62 11.795 40 180.39

0.30 3.910 20 11.793 44 180.32

3.2. Resistance analysis and MR of YBFNO

Fig 4 shows curves in logarithmic scale between resistance and temperature of YBFNO samples in 0T and 1T magnetic fields. Samples of YBFNO in the two magnetic fields remain semiconducting like YBFO. It was reported that the resistivity function of LaBa2FeO8.3 (150 K＜T＜300 K) acted as Arrhenius-type behavior with an activation energy of about 0.2 eV [6]. Elzubair et al. [7] formulated an evident variable-range hopping (VRH) process:

( )1 40exp T Tρ ρ∞= (1)

46.0 46.5 47.02θ (°)

Inte

nsity

(cou

nts)

x=0.00

x=0.05

x=0.10

x=0.15

x=0.30

x=0.50

x=1.00


Where is characteristic temperature related to the density of state at the Fermi level. The VRH process was suitable for describing the resistivity of RBa

0T

Fe O8+w2 3 (R=La, Nd, Sm, Eu, Gd) when measurement temperature range was increased to 620 K [8]. We find that the VRH process also fits for YBFO when measurement range was 10–300 K, and there was a direct link between the resistivity and temperature. Comparing fig 4(b) with fig 4(a), we notice limited effects of application field on the resistivity of YBFNO.

To compare the density of states at the Fermi level, we investigated the relationship between the resistivity and the Ni content in YBFNO at 300 K. The curves in fig 5, as a whole, are very similar to two curves in 0 T and 1 T magnetic fields at 300 K, except for x=0.15. It is shown that except for x=0.15 at 300 K, the application field has limited effects on the resistivity. Moreover, when Ni content is 0, 0.15, or 1, the resistivity of samples in 0 T and 1 T fields at 300 K is noticeably different from other samples, as is probably related to the onset of spatial charge or magnetic ordering.

Fig 6 shows the relationship between the resistance and the temperature for different Ni-content samples in 1 T field. With the decrease in temperature, the crystals experience Insulator-Metal (IM) phase transition.

256 K 123.46 K 66.64 K

0.25 0.30 0.35

20

15

10

5

0

T -1/4 (K-1/4)

Fig 4. Representative logarithmic resistance of YBa2Fe3-xNixO8+w in 0 T (a) and 1 T (b) fields. The continuous line is a fit to a VRH process. For convenience, only several lines were given

Fig 5. The relationship between the resistivity and Ni content in YBFNO at 300 K

0.0 0.2 0.4 0.6 0.8 1.0

20

(×

103 Ω

·cm

)

x

0 T1 T

16

12

8

4

0

0.25 0.30 0.35

20

15

10

5

0

Log e

R (Ω

)

T -1/4 (K-1/4)

x=0.00 x=0.05 x=0.10 x=0.15 x=0.30 x=0.50 x=1.00

66.64 K 256 K 123.46 K

Log e

R (Ω

)

x=0.00 x=0.05 x=0.10 x=0.15 x=0.30 x=0.50 x=1.00

(b) (a)


The temperatures of this phase transition TIM (nearly 110 K) are much lower than the Neel transition temperature, TN (nearly 660 K for pure YBFO). This is because TIM is closely relevant to the factors such as synthesizing conditions, crystal boundary, and magnet ordering. Moreover, the oxygen content in crystal also plays a role in the low TIM.

Fig 6. The relationship between the resistance

and the temperature of YBFNO in 1 T field Fig 7. The relationship between the MR and the

temperature of YBFNO samples in 0 T field

Fig 7 shows the relationship between the magnetoresistivity (MR) and the temperature. MR is obtained by

( )1 40exp T Tρ ρ∞= (2)

Where 0ρ and Hρ are resistivities in 0 T and 1 T fields, respectively.

The different intensity peaks corresponding to x=0.1, 0.15, 0.3, and 0.5 appear within a low-temperature range of 90–130 K and a high-temperature range of 240–300 K.

This phenomenon can be attributed to the effects of Colossal Magneto resistance (CMR). The maximum positive MR value at x=0.1 reaches nearly 600%. Other CMR peaks in the low-temperature range are flat and small, because of the phase transition between Paramagnetic-Metal (PM) and Ferromagnetic-Insulator (FI), instead of the one between PM and Ferromagnetic-Metal (FM). Impurities such as Ba2Fe2O5 and BaFe2O4 are rarely seen in the synthesized products. In addition, phase transitions probably exist between PM and FM at x=0, 0.05, and 1 within the both temperature ranges.

IV. ENDING REMARKS

We used the wet chemical method to prepare YBFNO samples. A small amount of Ni does not tremendously transform the space structure of YBFO system. The main reason for the decrease of lattice parameters is that Ni has smaller ionic radius than Fe. In YBFNO, Ni can

100

MR

(%)

T (K) 200 300

600

200

0

−200

x=0.00 x=0.05 x=0.10 x=0.15 x=0.30 x=0.50 x=1.00

400

100

R (×

108 Ω

)

T (K)

200 300

6

4

2

0

x=0.00 x=0.05 x=0.10 x=0.15 x=0.30 x=0.50 x=1.00


arbitrarily substitute partial Fe in the parent phase. YBFNO samples exhibit the semiconducting feature like YBFO. The VRH process is suitable to describe the resistivity of YBFNO samples.

ACKNOWLEDGEMENTS

The work was supported by Funds from Center University of the Education Ministry

(No. SWJTU09ZT24).

References

[1]. M.E. Massalami, A. Elzubair, H. Ibrahim, et al., Structural and magnetic properties of YBa2Fe3O7+x Fe-

based 1-2-3 ceramic oxide, Physica C, 1991, 183(1-3): 143-148.

[2]. Q. Huang, P. Karen, V.L. Karen, et al., Neutron-powder-diffraction study of the nuclear and

magnetic structures of YBa2Fe3O8 at room temperature, Phys. Rev. B: Condensed Matter, 1992, 45(17):

9611-9619.

[3]. Q. Huang, P. Karen, V.L. Karen, et al., Neutron-powder-diffraction study of the nuclear and magnetic

structures of the substitution compound (Y Ca1-x x)Ba2Fe3O8+y (x=0.05,0.10,0.20), Phys. Rev. B: Condensed

Matter, 1994, 49(5): 3465-3472.

[4]. P. Karen, A. Kjekshus, Q. Huang, et al., Neutron powder diffraction study of nuclear and magnetic

structures of oxidized and reduced YBa2Fe3O8+w, Journal of Solid State Chemistry, 2003, 174(1): 87-95.

[5]. Q.Z. Huang, V.L. Karen, A. Santoro, et al. , Substitution of Co3+ in YBa2Fe3O8, Journal of Solid State

Chemistry, 2003, 172(1): 73-80.

[6]. A. Elzubair, M. ElMassalami, Thermally activated electronic hopping and oxygen nonstoichiometry in

the Perovskite LaBa2Fe3Ox, Physica B, 1996, 225(1-2): 53-62.

[7]. N.F. Mott, E.A. Davis, Electronic Processes in Non-Crystalline Materials, Oxford: Clarendon Press, 1971.

[8]. A. Elzubair, M.E Massalami, P.H. Domingues, On the structure and magnetic properties of the series

RBa (R=La,Nd,Sm,Gd), Physica B, 1999, 271(1-4): 284-293♦ 2Fe3O8+x


ECONOMIC COMPETITIVE EVALUATION OF HIGH-SPEED LINES

PH.D. KURENKOV P.V Department "Transport business" MIIT POLYAEVA T.I Department "Economics and Logistics in the transport" SamGUPS

Motion with speeds above 250 km/h at the moment is a priority for the strategic development of rail transport from 23 countries worldwide. High-Speed Lines (HSL), which are considered globally competitive in relation to air transport for distances up to 700 km, in last years have tended to the length with increasing speed. To Russia, even where in the European part of the way a distance between cities with a population of five hundred thousand to half a million people are at the length of 700 km the problem of building and recovery of such "extended" HSL has particular actual.

The first HSL in the world had a length of 570 km (Tokyo - Osaka), and in Europe, 470 km (Paris - Lyon). At these days, the longest in the world of HSL is constructed in China between Beijing and Shanghai (1,320 km), the presumed value of $ 32 billion. High-speed passenger traffic (HST) on its scheduled non-stop, because of what will be achieved a high-speed route. At the moment, night trains on this route also followed without any interruption. Primarily the demographic situation serves in China. Beijing (19.7 million people) and Shanghai (24.6 million people) are the largest metropolitan China. Accordingly, the passenger traffic flow between the Chinese capitals greater than that between the Russian cities.

Table 1. Dynamics of carriage passengers by HSL in Europe (1994 - 2004) according to the UIC/3, p. 31/

France Germany Italy Spain Rest of Europe Europe Year Pass-km

(millions) Growth

(%) Pass-km

(millions) Growth

(%) Pass-km (millions)

Growth (%)

Pass-km (millions)

Growth (%)

Pass-km (millions)

Growth (%)

Pass-km (millions)

Growth (%)

1994 21,9 - 8,2 - 0,8 - 0,9 - 0,3 - 32,1 - 1995 21,4 - 2,3 8,7 6,1 1,1 37,5 1,2 33,3 0,4 33,3 32,8 2,2 1996 24,8 15,9 8,9 2,3 1,3 18,2 1,3 8,3 1,4 250,0 37,7 19,9 1997 27,2 9,7 9,3 4,5 2,4 84,6 1,5 15,4 2 42,9 42,4 12,5 1998 30,6 12,5 10,2 9,7 3,6 50,0 1,5 0 2,7 35,0 48,6 14,6 1999 32,2 5,2 11,6 13,7 4,4 22,2 1,7 13,3 2,8 3,7 52,7 8,4 2000 34,7 7,8 13,9 19,8 5,1 15,9 2,2 29,4 3,5 25,0 59,4 12,7 2001 37,4 7,8 15,5 11,5 6,8 33,3 2,4 9,1 3,8 8,6 65,9 10,9 2002 39, 6,7 15,3 -1,3 7,1 4,4 2,5 4,2 4 5,3 68,8 4,4 2003 39,6 -0,8 17,5 14,4 7,4 4,7 2,5 0 4,1 2,5 71,1 3,4 2004 41,5 4,9 19,6 12,0 7,9 6,6 2,8 9,9 4,1 0 75,9 6,8

In France, a prerequisite for the construction of SCM also served as a congestion of railway line Paris - Lyon (410 km). Design of the line, called the South-East, was completed by 1981. Reduced travel time (about 3 hours) caused an unprecedented increase in the number of passengers carried. It was a commercial success that has inspired many countries to expand or


built a high-speed rail network.

High-speed rail is more profitable than usual, not counting the cost of constructing infrastructure. Reason - the fact that a lot of operating costs such as a staff have a fixed cost per hour, while the income from ticket based on the distance. Passengers also pay more for a high speed per mile. Thus, the ratio of operational revenue/cost more for high-speed systems.

Fig 1. The cost of construction of 1 km of the new HSL /3, p.25/

At the moment we have an obvious contest in the development of high technology between the European and Asian countries. And if previously the undisputed leader in this area were the Europeans, in the long term for the year 2025 in Asia will be built - 21 452 km of HSL, while in Europe will be built 17 565 km high-speed lines. Major Trends in construction and operation of BCM in the world are displayed in fig. 1.

Obvious results the acceptance actively implemented introduce in the context of continental China program for the construction of high-speed lines. As of 2007, the Shenyang-Qinhuangdao line pro ¬ pulling over 400 km of secondary and technical train speed was 197.1 km/h. From ¬ covering August 2, 2008 ¬ HST Beijing-Tianjin line length of 117 km, calculated at a maximum speed of 350 km/h, reduced the time the train between the two cities from 70 to 30 minutes. The highest value recorded ¬ Noah technical speed to the mainland in 2009 amounted to 236 km/h. This was achieved in the family rides CRH, developing a maximum speed of over 300 km/h / 4 /.

0

2000

4000

6000

8000

10000

12000

14000

China Taiwan India Iran Japan Saudi Arabia

North Korea

Turkey

planned lines

lines under construction

operating lines

Fig 2. Planned, built and operated by HSL Asia km /5/


At the moment, the longest in the world of SCM has built in China between Beijing and Shanghai (1,320 km), the presumed value of $ 32 billion.

0

1000

2000

3000

4000

5000

6000

planned lines

lines under construction

operating lines

Fig 3. Planned, built and operated by HSL Europe, km

Currently in Europe the process of connecting high-speed systems of different countries into one network TEN-T, which is part of the sustainable spatial development of the European continent. The situation here is exactly the opposite of China. HSL connects densely populated areas, such as Corridor 6 MAF Lyon (1,5 million people) - Turin (0.9 million people) - Trieste (0.2 million people) - Ljubljana (0.3 million people) - Budapest (1, 7 million) in length about 1500 km. The average length of high-speed lines in Europe is 300 km, which a priori ensures competitiveness and a high return on rail transport. The average distance between the cities of one million people in Europe is 200 km. Therefore, competition between operators of high-speed lines and airlines on intra-European market low-kosters rather difficult.

Now in Europe passengers preference high-speed trains near 60 to 85% of. And, for example, on the lines of the Paris-Brussels-Cologne and Frankfurt are completely displaced airlines. True, the success of HSL, not least ensured by an intensive development of intermodal transport, where the airline and railroad companies are moving from direct competition to cooperation, giving passengers the opportunity to travel on one ticket. In this case, the medium-and long-haul travel by air require high-speed traffic on the initial and final sections of the path to deliver passengers to the airport and back.

Many operators do not think that more speed achievement is main. In most cases, an increase of frequency of trains is more important in terms of attractiveness for passages as, for example, in the UK.

Currently, there are a lot of the concepts of "high-speed traffic" (IRR) on a rolling stock, and on infrastructure (road, signaling, centralization and blocking, electrification and power supply, etc.), as well as on the compatibility of rolling stock and the infrastructure.

In terms of the Directive 96/48/E performance enough to cover the entire rail infrastructure, capable to provide high-speed services. However, in practice the speed is not always an indicator that can clearly delineate the boundary between the speed and high-speed traffic. In many countries, the commercial rate is limited by factors such as passage of a densely populated urban areas (to be slowing down to reduce noise levels and the risk of accidents), or because of


the use of tunnels and viaducts (where the speed limit is set to 160-180 km / h from for security reasons).

In Europe, there are two types of countries - France and Spain have lower cost of construction of 1 km road than Germany, Italy and Belgium. The reason is not only geography, but also the construction of the path used in the construction of the HSL. So in France, construction costs are minimized by eliminating the tunnels and viaducts. Due to the fact that in France the HSL is used only for passenger traffic, slope may be 3.5%, while for the mixed type - only 1-1.5%.

Table 2. The cost of operating the infrastructure 1 km single-track section of HSL in Europe according to UIC in 2008 / 3, p.26 /

Belgium France Italy Spain 142 km 2638 km 492 km 949 km

Rolling stock 13 834 43,7% 19140 67,3% 5941 46,0% 13531 40,4%Energy 2576 8,1% 4210 14,8% 2455 19% 2986 8,9%Alarm 3248 10,3% 5070 17,8% 4522 35% 8654 25,9%

Telecommunications 1197 3,8% 0 0 0 0 5637 16,8%Other costs 10821 34,2% 0 0 0 0 2650 7,9%

Total 31683 100% 28420 100% 12919 100% 33 457 100%

In all cases, the cost of operating the way has a share from 40% to 67% of total expenditure, while the share of only 10-35% of signaling, which is less than the usual ways (15-45%).

With respect to interaction with the traditional railway network of HSL appropriate to classify as follows:

- Mixed traffic (new or reconstructed line). In this case high-speed trains also go to the usual lines. Mixed traffic is dominated mainly by German, Spanish and Italian high-speed lines, as well as in the UK, Austria, Switzerland, the Netherlands, Belgium, Finland and China.

- Only high-speed traffic with access to a regular network (mostly new line). In this case, high-speed lines runs only specialized rolling stock, which often comes at the same time and on a regular network, especially in the areas of hubs and in the final sections of routes. Typical examples are the French line.

- Stand-alone high-speed network communications (only new lines). Here, the transfer of rolling stock between conventional and high-speed networks is not provided. An example is a network of high-speed lines in Japan.

- Mixed traffic on a standard European track 1,435 mm with access to the local track by having a rolling of the wheel sets. The Spain railways of Spain's gauge of 1668mm (narrow-gauge lines at 1000 mm). An example of this is the high-speed rail system in Spain AVE, a country with advanced technologies in the gauge change. Automatic points change wheelsets first, second and third generation networks combine various gauges. Daily through the points of change wheelsets are about 60 trains. Since 1969, these items are served over a quarter of a million trains.


At the moment, ready to introduce automatic gauge change system of the 4th generation. Unlike previous systems, which apply only to passenger traffic, the new system would:

- Automatic change of gauge passenger and freight trains; - The use of rolling stock for different manufacturers; - The ability to change the track width of mixed trains, that means trains with different

types of cars and manufacturers. The differences between the systems can be displayed graphically HS systems (see Fig.

4.) For the adopted distance of 250 km length, which is the usual train passes for 2 - 2.5 hours. Time frame for which the same distance will be high-speed trains of different types of HSL depends on the technical configuration, the values of slope, the number of arrows, viaducts, etc.

Geographical expansion and range requirements for the product in use meaning ¬ increase in the number of types and series of high-speed trains. Most important trends in their development are:

• Achieving a maximum velocity ¬ velocity of 350 km/ h, and (above all in Europe) to ensure exploitation compatibility with infrastructure of the railways in various countries every ;

• An increase in passenger capacity, while maintaining an adequate level of comfort; • Reducing energy consumption as energy the manufacture and in use; • Use of common construction platform to accelerate development and reduce costs in

manufacturing; • A high level of modularity and interchangeability for reduction of costs.

Fig 4. The position of different systems of SCM and conventional rail

in relation to time intervals and distance / 3, s.22 /

After 2012 it is expected that the Asian demand for new rolling stock will decline due to the completion of the program complete replacement of the Japanese high-speed trains of first generation. In contrast, in Western Europe the market will continue to grow as the replacement program trains first generation in France and Germany are in the initial stage. Another confirmation of this tendency is a program may be for, which announced in the CIS countries, South America and North ¬ of South Africa / 1, 2 /.


Table 4. Types of high-speed trains in the world /3, p.27/ Country Type of Train Year Seats

number Serviced distance

(km) Maximal speed

TGV Reseau 1992 377 495 300/320 TGV Duplex 1997 510 525 300/320 France

Thalys 1996 377 445 300/320 ICE-1 1990 627 500 280 ICE-2 1996 368 400 280 ICE-3 2001 415 420 330

ICE 3 Polyc. 2001 404 420 330 Germany

ICE-T 1999 357 360 230 ETR 500 1996 590 360 300 Italy ETR 480 1997 480 288 250

Spain AVE 1992 329 470 300 CRH380A 2010 494 420 350/380 China CRH380AL 2010 1027 420 350/380

500 1995 1323 550 320 700 1997 571/1323 550 285 Japan

N700 2007 1323 550 300

Of particular importance is the account of expenses for operation and maintenance of trains. In the world of every country that has so far developed its own HST type of rolling stock, the most responsible realities of life and needs (see table. 4.).

Table 5. HSL technology in Europe: Comparing the costs of operation and maintenance according to UIC (2008)/3, p.28/

Operating costs Maintenance costs Country Type of

Train For train (mln)

For seat

For seat-km

For train (mln) For seat For seat-km

TGV Reseau 17,0 45,902 0,0927 1,6 4,244 0,008

TGV Duplex 20,8 40,784 0,0776 1,6 3,137 0,005 France

Thalys 24,8 65,782 0,1478 1,9 5,039 0,011 ICE-1 38,9 62,041 0,1240 3,1 4,944 0,009 ICE-2 26,0 70,652 0,1766 1,4 3,804 0,009 ICE-3 17,9 43,132 0,1026 1,6 3,855 0,009 ICE 3 Polyc. 20,4 50,495 0,1212 1,7 4,207 0,010

Germany

ICE-T 15,5 43,417 0,1206 1,8 5,052 0,014 ETR 500 34,1 57,796 0,1605 4,0 6,779 0,018 Italy ETR 480 21,1 43,958 0,1526 3,2 6,666 0,023

Spain AVE 23,7 72,036 0,1532 2,9 8,814 0,018 Maintenance costs depend on the distance between the depot and the terminus, the average

waiting times at the depot, the time of shunting. Greater share of the cost of maintenance costs take on staff salaries. The flow rate depends on the technology used by the operator. The average cost per seat (see Table. 5.) Account for 53 €. For AFL 500 km long, considering the


duty cycle trains 100%, the average operating cost per passenger ranges from 41,3 € (2000) for the TGV Duplex, up to 93 € for the ICE-2 in Germany.

In Russia, is an issue of competitiveness and profitability of high-speed projects in most of the territory. Long distances between major conurbations and sparsely populated suggest only a network version of HSL. And the main principle is to reach the largest possible number of densely populated areas. This principle applies to the example of the proposed route Moscow - Nizhny Novgorod - Kazan (with a branch in Samara) - Ekaterinburg. Distance along the branch Samara - Kazan - Nizhny Novgorod - Moscow 1300 km, and the branch Samara - Penza - Moscow 1030 km, much less with the high cost per km of new road. The population of Kazan, $ 1.1 million, Nizhny Novgorod - 1.3 million people. In major cities at an alternate route home: Syzran - 0.2 million people in Penza - 0.5 million people in Ryazan - 0.5 million people.

Primarily based on return line is the proximity of highways to a greater number of consumers, so the advantage is on the side of more extended, but more densely populated route.

If you continue to branch in Samara to Saratov and Astrakhan, or through Samara to Orenburg is possible to connect the Volga and Southern Urals into a single high-speed network. In addition, the route Samara - Orenburg is a perspective of international importance and is indirectly linked to China's ambitious plans for the organization of high-speed traffic through Beijing - London. The most likely way for Transeuroasian route lies on the line Beijing - Hohhot - Hoto - Bao - Tou - Uday - Urumqi - Friendship - Aktogai - Astana and then through Russia en route Orenburg - Samara - Penza - Moscow, where it is possible to extend the line in the directions of Brest - Kosice - Budapest - Rome, Warsaw - Prague - Munich and so on.

Network organizing principle of HSL in Russia will intensify passenger traffic and economic activity not only in the direction of the center - periphery, but also between regional industrial and economical centers.

Economic evaluation of infrastructure upgrades, construction of new facilities, changes in technology, transport processes and other activities for the organization of the HST is the following: cost accounting HST; economic evaluation of changes in the schedule and timetable of the freight train, options for local work stations and areas in the implementation of the HSL, the options interaction of urban, trunk, air and river passenger transport options for the high-speed lines: the construction of special ramps, construction of new railway lines, the upgrading of existing lines under the HSL.

References [1]. Высокоскоростное движение: что впереди? // Железные дороги мира. – 2009. - 7. – С. 9-20. - Rus. [2]. Высокоскоростные сообщения: частота важнее скорости // Железные дороги мира. – 2010. - 3. – С. 9-22. - Rus. [3]. Rus G., Barron I., Nash C, Economic Analysis of High Speed Rail in Europe // Fundacion BBVA. – 2009. – P. 140. [4]. Материалы сайта http: // rbcdaily.ru[5]. Материалы сайта www.hsrail.ru♦


http://rbcdaily.ru/

CALCULATION PROCESS OF SETTLEMENT - TRANSITION PILE NET FOR BRIDGE APPROACH EMBANKMENTS

PhD. VU THE SON University of Transport and Communications MA. NGUYEN VAN SUU Transport Engineering Design Joint Stock Incorporated South MA. TRAN HUU BANG Sai Gon Construction Quality Control Joint Stock Company

Summary: Based on the analysis of load transfer structure in pile net, this report summarizes the calculation content of pile net solution in the settlement transition treatment at bridge approach embankments on soft soil. Calculation content includes: Determine the length of pile net, calculate the vertical forces on the pile caps, the extreme load capacity of piles, settlement of pile and geosynthetic. The calculation example is applied for the Abutment A2- Binh Loi Bridge of Tan Son Nhat - Binh Loi - Outer Ring Road Project with the necessary Pile net length to ensure both stability and the slide is 24m. The calculated pile lengths vary from 22m to 12m, from the abutment adjoinment to the end of the pile net. Corresponding settlement after 15 years remaining at the end of the pile net is 17.5cm.

I. INTRODUCTION

The pavement of bridge approach embankments is often settled, cracked and broken at the start and the end of the approach road. In Vietnam, as a result, vehicles cannot run smoothly and cannot reach high-speed after an operating time of load transfer platform. This affects the operation efficiency of roads. There are many solutions of soft soil treatment in approach road with high efficiency such as: Prefabricated vertical drain, Sand drain, Deep cement mixing, Pile slab [8]. However, these solutions have not completely solved the above matters.

It is important to consider, evaluate and find a solution of soft soil treatment for approach roads that ensures the smooth run of vehicles and maximizing the operation capacity of roads.

Based on the analysis of behavior of the pile net and soil, as well as related documents, foreign and domestic specifications; this paper proposes calculation content of settlement-transition pile net. The calculation content of pile net solution is illustrated through the example of Abutment A2 at Binh Loi Bridge in Tan Son Nhat - Binh Loi - Outer Ring road Project.


II. PILE NET SOLUTION FOR SETTLEMENT TRANSITION

Geosynthetic combining with the reinforced concrete piles with variable lengths is known as settlement - transition pile net as shown in figure 1 and figure 1.

Figure 2. Typical cross section of Pile net solution

Piles adjacent to abutment are calculated for fully length restrained in hard soil layer. The length of remaining pile increases from pile net to abutment position, so settlement is also reduced correspondently to create a settlement transition area.

Figure 2. Profile of Pile net solution of settlement transition

In some countries, geosynthetic has been placed on piles (reinforced concrete piles, deep cement mixing) with a constant length to treat the bridge approach road.

III. CALCULATION CONTENT

Based on the analysis of load transfer in Pile Net, report proposes calculation content of pile net solution for settlement transition as following figure 3:

Figure 3. Calculation content of pile net solution for settlement transition

3.1. Determine the length for Pile Net

Assess the stability of pile net vertical road method based on circular slip through Bishop GeoStudio 2004 software to audit [3]. Based on the sliding characteristics of the arc length round to define the length for pile net


3.2. Load transfer mechanism in Pile Net

It is important to consider the load transfer mechanism in Pile Net from the interaction between the piles, geosynthetic and soil. This is the basis issue for calculation of settlement-transition pile net.

Considering the load distributing on Pile Net as figure 3 [7]:

Figure 4. Load transfer structure in Pile Net

With: Ws - Surcharge effect on embankment (kN/m2); W1 - Weight of embankment between the adjacent pile caps (kN); W2 - Weight of embankment on pile caps (kN); Wt - The vertical load effect on the strengthened reinforcement between the adjacent pile caps (kN/m2); P’

c - The vertical stress on pile caps (kN/m2); Trp - The tensile load in the reinforcement (kN/m); τ - Shear resistance of soil (kN/m2); H - Height of embankment (m); a - The size of the pile caps (m); S - The spacing between the adjacent piles (m).

Strengthened reinforcement combine with good soil layer cover above, has effect transferring the embankment load between piles to pile caps. Blocks of embankment will be moved down and settlement much because the below soil layer of Strengthened reinforcement is very weak. Transition is reduced by shearing resistance of embankment on the pile cap. Shear strength of embankment on the pile cap reduce the pressure on Strengthened reinforcement but it make increase the pressure on the pile cap.

To calculate Pile net, It is necessary to calculate the value of vertical pressure on the pile cap. The order of calculation is as following:

- Define the average vertical tress at the base of the embankment follow formula [6]:

(1) 'v fs sσ = DL.f + W .fq

With: DL - Dead load effect on piles (T/m2); ffs - The partial factor for soil unit weight (refer to table [6]); fq - The partial load factor for surcharge (refer to table [6]).

The vertical stress on the pile caps is determined as following formula [6]:

c

2

' cv

C .aP = .σ

H⎡ ⎤⎢ ⎥⎣ ⎦

' (2)

With: Cc - The arc coefficient.


3.3. Calculation of the vertical forces on the piles cap

Calculation of the vertical forces on the piles cap are determining load (Dead load and live load) on piles cap. The vertical forces on the piles cap calculated follow formula [5]:

(3) ' 2a cQ = P .a

3.4. Calculation of total pile capacity

Total pile capacity include is resistance of pile (Skin friction and end bearing).

- Extreme Load of pile according to soil is calculated by the formula [3]:

(4) u pQ = Q + Qs

With: Qp - Resistance of pile cap (ton); Qs - Resistance of pile side (ton).

- Safety factor about Load ability of Pile is calculated as following [2]:

us

a

QF =

Q (5)

3.5. Calculation of Pile settlement

Method of settlement analysis of pile groups in Pile Net similar to methods used in common pile foundation [2].

Settlement estimation method is according to Guideline at Section VI of survey design process of embankment on soft soil layer 22 TCN 262-2000 [1].

3.6. Calculation of reinforcement

Reinforcement is located below embankment to increase the strength of embankment, avoid embankment being destructed by excessive deformation or sliding cutting into below soft soil layer.

According to British Standard BS 8006:1995 when designing strengthen reinforcement in this case according to the theoretical of flexible cable system [6].

a. According to longitudinal Direction of embankment

Considering calculation diagram as figure 5:

Figure 5. Calculation diagram of tension with longitudinal direction of embankment


With: d - Diameter of pile (m); Trp - Tension of reinforment with longitudinal direction of embankment (kN/m).

Tension Trp in reinforcement is determined as follow:

Trp

W (s - a) 1T = 1+2a 6ε

(6)

With: ε - The elongation of strengthened reinforcement according to rolling direction.

Direction along the embankment length, maximum tension is necessary strength to move up the vertical load on the pile cap Trp.

(7) r rT = T p

b. Horizontal direction of embankment

Considering Calculation diagram of strengthened reinforcement as figure 6:

Figure 6. Calculation diagram of tension with horizontal direction of embankment

With: Tds - Inner Tension of reinforcement with horizontal direction of embankment (kN/m); Le - Minimum length of anchor of reinforcement with horizontal direction of embankment (m); Lb - The reinforcement bond length needed beyond the outer row of piles across the width of the embankment (m); Lp - The horizontal distance between the outer edge of the outside pile cap and the footing of embankment (m).

Tension Tds of strengthened reinforcement need to determine as following:

ds a fs q sT = 0.5K (f .γ.H + 2f .W )H (8)

With: Ka - The coefficient of active soil pressure; γ - The unit weight of the embankment material (kN/m3).

According to the horizontal width of embankment, the biggest tension must be the total of necessary strength to transfer the vertical load on the pile cap Trp and necessary strength to resist the horizontal thrust Tds.


(9) r rpT = T + Tds

IV. EXAMPLE

4.1. Data for calculation

- The physic-mechanical property of this layer at Abutment A2 - Binh Loi bridge of Tan Son Nhat - Binh Loi - Outer Ring Road Project [4].

- Unit weight of embankment Material [3]: γađ = 2.3 T/m3, γsb = 2.0 T/m3, γs = 1.8 T/m3, γc = 2.4 T/m3;

- Data for calculation of reinforcement [3]: a = 1m, Bemb = 29m; n=1.5; Hfill =3.69m; φcv = 300, Ws = 1.18 T/m2

.

4.2. Calculation results

Summary results is displayed as Table 2 and Settlement is shown on diagram 7 [3]:

Figure 7. Settlement transition of pile net

Remark: according to diagram, it shows that the settlement increases from the position close pier to the end of pile net. For transiting settlement, calculate the residule settlement after 15 years of normal embankment section same with the settlement of the last pile when starting at pile net.

Calculation results of reinforcement is showed in table 1 [3]:

Table 1. Calculation results of reinforcement

Item Units Results

The tensile load in the reinforcement kN/m Trp = 22

The tensile load in the reinforcement (prevention of lateral sliding) kN/m Tds = 208

Maximum tension in reinforcement kN/m Ttot= 257

Extreme tensile stability of reinforcement kN/m Tult= 55


Remark: Based on the calculated results, the appropriate type of geosynthetic (according to specification of manufacturers) must have the biggest pull-out strength along longitudinal direction of Tult = 55 kN/m and along horizontal direction of Ttot = 257kN/m.

Table 1. Summary of calculation result about Pile Net of Settlement Transition

Row No 1 2 3 4 5 6 7 8 9 10 11 12 13

Distance from abutment (m)

1.2 3 4.8 6.6 8.4 10.2 12 13.8 15.6 17.4 19.2 21 22.8

Height embankment (m)

3.69 3.64 3.58 3.52 3.47 3.41 3.36 3.31 3.26 3.21 3.16 3.10 3.05

Pile length (m) 22 22 22 20 20 20 18 18 16 16 14 14 12

Spacing (m) 1.2 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8

Skin friction (ton)

65.1 65.1 65.1 54.9 54.9 54.9 46.2 46.2 39.1 39.1 28.3 28.3 17.4

End bearing (ton)

27.5 27.5 27.5 26.2 26.2 26.2 16.7 16.7 5.3 5.3 5.3 5.3 5.3

Total pile capacity (ton)

92.6 92.6 92.6 81.1 81.1 81.1 62.9 62.9 44.4 44.4 33.5 33.5 22.7

The vertical stress on the pile caps (ton/m2)

24.2 24.2 24.2 23.3 23.3 23.3 22.5 22.5 21.9 21.9 21.4 21.4 20.8

The vertical forces on the pile caps (ton)

24.2 24.2 24.2 23.3 23.3 23.3 22.5 22.5 21.9 21.9 21.4 21.4 20.8

Safety factor 3.8 3.8 3.8 3.5 3.5 3.5 2.8 2.8 2 2 1.6 1.6 1.1

Total settlement(cm)

0 0 0 1.3 1.3 1.3 2 2 5.6 5.6 17.5 17.4 26.0

Settlement after 15 year (cm)

0 0 0 1.3 1.3 1.3 2 2 5.4 5.3 14.2 14.1 17.5

Consolidation (%)

0 0 0 100 100 100 100 100 96 96 81 81 67

Remark: To calculate the settlement transition at abutment A2 - Binh Loi Bridge, piles adjacent to abutment are calculated for fully length restrained in hard soil layer. The length of remaining piles increases from the starting position of pile net to abutment (Lpile = 12 ÷22m), therefore the settlement is also correspondently reduced (S15year = 17.5÷0 cm).


V. CONCLUSION

This paper proposes the calculation content of pile net solution in the settlement transition

of bridge approach embankment including: Determine the length of pile net, calculate the

forces distribution on piles cap, the extreme load capacity of piles, settlement of piles and geosynthetic. The length of piles decreases from the starting position of bridge approach road to abutment to solve problem of settlement transition. Piles adjacent to abutment are calculated for fully length restrained in hard soil layer. The length of piles at the end of the pile net is calculated so that the remaining settlement after 15 years is equal to that value of common embankment section to create the area of settlement transition.

The application is for the Abutment A2- Binh Loi Bridge of Tan Son Nhat - Binh Loi - Outer Ring Road Project. Pile lengths vary from 22m to 12m from the abutment to the end of pile net. Corresponding settlement after 15 years remaining at the end of pile net is 17.5 cm. The selected geosynthetic should be able to bear the biggest pull-out strength longitudinally of Tult = 55 kN/m and horizontally of Ttot = 257 kN/m.

References

[1]. Ministry of Transport. 22 TCN 262 - 2000 Standard. The publisher transport and communication, Ha Noi, 2001.

[2]. Ministry of Construct. Pile foundation - TCXD 205 - 1998 Standard.

[3]. Nguyen Van Suu. Applied research of pile net solution in the settlement transition treatment at approach road embankment on soft soil. Master thesis of scientific and technical, University of Transport and Communications, 2011.

[4]. Tedisouth. The report geological of Tan Son Nhat - Binh Loi - Outer Ring Road Project. Technical design, 2009.

[5]. Braja M. Das. Principles of Foundation engineering, Third Edition. California State University, Sacramento.

[6]. British Standard BS 8006: 1995. Code of Practice for Strengthened/Reinforced Soils and Other Fills. British Standard Institution, London.

[7]. Rutugandha Gangakhedkarga. “Geosynthetic Reinforced Pile Supported Embankment”. Master of Engineering, University of Florida, 2004.

[8]. Anand J. Puppala. Recommendations for design, construction, and maintenance of bridge slabs. Synthesis Report, the University of Texas, 2009♦


ADVANCES IN DESIGN THEORIES OF HIGH-SPEED RAILWAY BALLASTLESS TRACKS

XUEYI LIU PINGRUI ZHAO FENG DAI Southwest Jiaotong University, Chengdu 610031, China

Summary: The design theories of the ballastless track in the world are reviewed in comparison with the innovative research achievements of high-speed railway ballastless track in China. The calculation methods and parameters concerning train load, thermal effect, and foundation deformation of high-speed railway ballastless track, together with the structural design methods are summarized. Finally, some suggestions on the future work are provided.

Key words: High-speed railway; ballastless track; design theory.

I. INTRODUCTION

Structure forms and design theories of ballastless tracks vary across the world due to the different development backgrounds. In Japan, the slab track was typically laid on the solid foundation such as a bridge or tunnel at first, and then gradually developed to the soil subgrade afterwards. It adopts the unit design that takes into account the effect of train load. The German ballastless track was first laid on the soil subgrade and then on the foundation of bridges and tunnels. Its continuous structure involves the consideration of thermal effects. The early ballastless track in China was mainly laid in tunnels with the chief concern being the influence of train load. With the increasing application of ballastless track, a relatively general design theory and a structural system have been gradually formed after the innovative research with high-speed railway ballastless track.

This paper reviews the calculation methods and parameters as well as the structure design procedures, and briefly introduces the advance in the design theories, of ballastless track based on the innovative research achievements in China. Finally, some suggestions on the future work are provided, including fatigue properties under the coupling action of train and temperature load, durability, long-term dynamic properties, and maintenance mechanics of the ballastless track.

II. OVERVIEW OF BALLASTLESS TRACK DESIGN THEORIES

In the design of Japanese slab track, the train load effect is a primary concern. Using the elastic design method, the security during the manufacturing, hoisting, and constructing of the slab track is


maximized. As seriously damaged CA mortar at the slab corner and the slab warping caused by temperature gradients emerged, the uneven support caused by warping is considered in the analysis [1]. In the baseplate design, in accordance with the limit state method, the train load and the subgrade’s uneven settlement are considered together with the influence of weather conditions, concrete contraction, and construction.

German developed its ballastless track by borrowing the design concept and method of pavement engineering [2]. Most has longitudinally continuous structure, and temperature load and concrete contraction are the main factors to be considered in the design. The reinforcement is located near the neutral axis and does not bear the train load. The effect of train load and temperature gradient is resisted by the rupture strength of the concrete.

In China, the early monolithic roadbed track, whose structure design mainly considers the train load, was applied in the tunnels with good foundation condition and little temperature variation. The structural design of the Suining-Chongqing railway took into account the effect of uneven foundation deformation and temperature load [3]-[4]. Following systematic research on the ballastless track, the design theory based on the allowable stress method was created with full consideration of train load, temperature, and foundation deformation effect.

In general, the design theory of ballastless track in different country was relevant to its own construction environment and structure evolution. The design theory proposed in different periods could meet the construction requirements for different types of ballastless track.

III. CALCULATION OF TRAIN LOAD STRESS

The track supports the train load and guides the vehicle operation. The calculation of train load stress must be considered in the ballastless track design. The elastic foundation beam model [5]-[6] is mainly used for calculation of the load stress in the traditional track structure. The model can be solved using the multilayer composite beam theory on the elastic foundation

[7]-[10] according to the complexity and analysis requirement of the track structure. In Germany, however, the Eisenmann theory [11]-[13] was adopted to calculate the stress of the rail structure under the train load. In this theory, rail is regarded as an infinite beam on the elastic foundation to calculate the support reaction of the fastener; the multilayer structure is translated into a monolayer one according to the connection status of the structural layer, and then the internal force and displacement of the converted monolayer structure under the action of fastener force is calculated using the infinite beam on the elastic foundation and Westgaard’s stress function.

To sum up, the main components are treated as flexural members in the train load design of ballastless track in China and Japan. This is because the ballastless track design was originally developed based on the traditional design methods for ballast track that put an emphasis on simulation of the force properties of main components and the generality of analysis method. In Germany, however, the design theory and parameters selection of ballastless track were developed from the experience of highway concrete pavement design; thus, its structural


difference in ballastless track can also be attributed to heritance of the traditional design theory.

In accordance with the structural characteristics that the rail and the sleeper are cross-supported on the elastic foundation in the ballast track, the cross beam model on the elastic foundation [14-15] was developed on the basis of the elastic foundation beam model, and can also be used for the stress calculation of the ballastless track [16] once the values of the model parameters are determined. Thanks to the development of the computing technology, the solid finite element model [17-19] can be employed to obtain the particular stress state inside the ballastless track structure.

As the major supporting structure of the ballastless track, the track slab (or bed slab) and baseplate (or supporting layer), whose deflections under the train load are far smaller than their thicknesses, have a far smaller size in the vertical direction than in the longitudinal or lateral direction. This feature conforms to the structural characteristics of the elastic plate. Consequently, the elastic plate [20] is generally adopted for simulation and analysis of the supporting structure of ballastless track. The rail, a slender structure, is reasonably simulated by the beam model, while the fastener and the intermediate elastic layer, as well as the foundation below, are simulated with different kinds of springs. As a result, a beam-plate model of ballastless track on elastic foundation [21]-[23] is built as shown in fig 1.

Kf

P P

ERJR

Es, hs

Eb, hb

kRDKi

Notes: ERJR is the flexural rigidity of rail, where ER is the modulus of elasticity of rail, and JR the moment of inertia of rail; Es and hs are the modulus of elasticity and thickness of track slab, respectively; Eb and hb are the modulus of elasticity and thickness of baseplate, respectively; Kf is the rigidity of fastener; Ki is the rigidity of intermediate elastic layer; kRD is the rigidity of foundation below; and, P is the train load.

Fig 1. The elastic foundation beam-plate model of ballastless track

The load stress of the track slab (or bed slab) and the baseplate (or supporting layer) in the longitudinal and lateral directions can be obtained by exerting a vertical train load on the rail. This avoids the calculation in the longitudinal and lateral directions separately in the multilayer elastic foundation beam model. Moreover, the computational accuracy [7] is higher than that via the composite beam model or the cross beam model, and the computing workload is less than that via the solid finite element model.

The design wheel load of the Japanese slab track takes into consideration the wheel load variation due to wheel tread damage and tolerates three times the static wheel load. In fatigue checking, the allowable wheel load is 1.45 times the static wheel load. On the basis of the


allowable value of the derailment coefficient, the design lateral force was determined, and the lateral force for fatigue checking takes half of the design lateral force. In the Germany ballastless track design, the load takes the UIC71 with the dynamic coefficient of 1.5 and the unbalance loading coefficient of 1.2. In China, the dynamic coefficient is based on the results of dynamic tests and simulation calculations of the ballastless track, and the design wheel load can be three times the static wheel load. Based on the design parameters and operation conditions of the ballastless track on the passenger dedicated line (PDL) in China, the coupling dynamics of train and track system is applied to the statistic analysis. Considering the construction and maintenance conditions of the ballastless track in China, it is suggested that the constant effect train load be up to 1.5 times the static load [24].

The Winkler foundation is used to support the ballastless track, and the diameter of the bearing plate has a significant influence on the foundation coefficient. The smaller the diameter, the larger the foundation coefficient [1]. However, when the diameter D is not less than 76 cm, the change in the diameter has little influence on the foundation coefficient. As for the ballastless track, the supporting area of the track slab or the supporting layer is relatively large. Thus, for simplicity, the trial value of the bearing plate with a diameter of 76 cm, namely k76, can be used for calculations. When the subgrade compaction capacity is represented by the deformation modulus, the layered elastic system mechanics [3] can be applied to analyze the displacement of the subgrade surface with the even load of the rigid bearing plate; thus, deducing the supporting rigidity of the subgrade surface [25].

Within every bearing layer of the ballastless track, the substructure is generally weaker than the upper structure, and may readily crack under the train load if plain concrete or cement stabilized materials are applied. Once cracking, the bending moment is not readily transferred at the crack location, resulting in a reduction in the entire rigidity and the modulus of elasticity. Therefore, the reduced elastic modulus is used for calculation [26]. As for the reinforced concrete structure, the reinforcement is helpful to improve the flexural rigidity of the structural layers. However, due to the possible cracking, the transmission of the bending moment at the cracked location may be weakened. Consequently, only the concrete elastic modulus is used for calculation, without consideration of the influence of the reinforcement and crack.

IV. CALCULATION OF TEMPERATURE STRESS

The ballastless track is exposed to the atmosphere. With changes in external temperature, the temperature in every structural layer will vary. Once the deformation of the ballastless track due to the changing temperature is restrained, the temperature stress will occur inside the structure. The ambient temperature variation with an effect on the ballastless track includes the yearly temperature variation and daily temperature variation. In addition, the contraction of concrete will cause distortion, which is equivalent to decreasing the temperature load acting on the concrete.

The design of the continuous ballastless tracks represented by Rheda, Züblin, and Bögl in Germany attach great importance to the temperature effect. In order to limit the width of the


temperature cracks within the admissible range and maintain the state of incomplete cracks [27], the ratio of reinforcement in the slab should reach 0.8%–0.9%, according to the German Ballastless Track Design Specification. As a result, the width of cracks is limited within 0.5 mm. From the viewpoint that the sum of the minimum stress of the reinforcement with the slab cracking and the bending stress increment under the dynamic load must be less than the reinforcement fatigue limit to guarantee the service life, it is supposed that the longitudinal ratio of the reinforcement must be larger than 1.0%, so as to meet the demands of crack width and service life.

The Japanese slab track design adopts unit structure, and temperature variation has little influence on the track slab. Thus, temperature effect is not considered in the design. Nevertheless, warping displacement of the track slab is found in tests, where the track slab is in a state of being incompletely supported. Therefore, to address the variation properties of the track slab due to temperature, a series of theoretical and experimental research has been conducted [4].

As for the continuous slab structure, under the action of concrete contraction and decreasing temperature, concrete may easily crack, causing a stress redistribution of the reinforcement and concrete inside the slab. In order to guarantee security and utility, it is necessary to control the reinforcement stress and crack width.

The continuous slab shows different stress and variation properties at various tension stages. Before the concrete cracks, the concrete deformation is coordinated by the reinforcement. When the tensile stress of the concrete reaches its tensile strength, it will crack and stop working, which leads to the bond damage adjacent to the cracks. At this moment, the plain section hypothesis does not fit any more, and the reinforcement at the crack location bears all the axial force. When the axial force increases to the yield strength of the reinforcement, the concrete is cracked severely without bearing the tension. All the axial force is born by the reinforcement, such that the reinforcement yielding becomes the limiting condition of the slab in tension. The cracking axial force of the continuous slab depends on the tensile strength of the concrete and the sectional area of the slab. The amount of reinforcement has little influence on the cracking axial force, while the ultimate bearing capacity completely depends on the yield strength and the area of the reinforcement. In order to avoid cracking, the minimum ratio of reinforcement of the continuous slab should be specified.

The cracking in the continuous slab go through two phases: incomplete cracking and complete cracking. At the stage of incomplete cracking, the amount of cracks increases with the increasing load, and the maximum crack width remains basically unchanged. At the stage of complete cracking, the number of cracks remains unchanged, while its width increases with the increasing load. In order to limit the crack width, the cracking should be controlled at the stage of incomplete cracking. In the cases of incomplete cracking, the maximum temperature force inside the slab depends on the tensile strength and the sectional area of the concrete. The temperature force calculated with the design tensile strength is regarded as the common temperature force (main force). And the temperature force calculated with the standard tensile strength is taken as the maximum temperature force for checking in design. Refs. [28-29]


elaborated the different expressions of fracture interval, cracking width and reinforcement stress at different stages of cracking, and the relevant design measures have been put forward.

As for the unit bed slab structure, the temperature force of the slab is influenced by the longitudinal resistance of the fastener at the top and the frictional resistance at the bottom, as well as the displacement limitation of the convex plate. The classification of the unitary and continuous structure depends on whether the temperature force leads to the full-section cracking of the slab.

V. CALCULATION OF WARPING STRESS

The external environment will affect the temperature and humidity of the concrete slab. The influence of external environment gradually weakens with the depth from slab surface. The uneven distribution of temperature and humidity inside the slab leads to its warping deformation. When the deformation is restrained by the bottom friction, dead load, stop blocks, and train load, the warping stress occurs.

According to the German railway code, it is hypothesized that the slab in the vertical direction has a linear temperature gradient of 50 /m. In the temperature field test of the ballastless track on Suining-Chongqing railway, the temperature gradient [30] of the track before laid is about 52.6–68.4 /m and the temperature gradient of the slab track in the longitudinal direction on the Jialingjiang bridge is approximately 40–80 /m [31], with a large dispersion, but all larger than that of 50 /m in Germany.

In terms of geography and climate conditions, China has severely cold areas, cold areas, and temperate areas. Referring to the recommended value of the temperature gradient in the field of highway pavement, in consideration of the structure characteristics of the ballastless track, we advise that the maximum positive temperature gradient of the uppermost structure of the ballastless track in China be 80–85 /m, 85–90 /m and 90–95 /m for severe cold area, cold area and temperate area, respectively, and that the temperature gradient distribute linearly in the vertical direction. The effect of temperature gradient can be neglected in the substructure. The negative temperature gradient can be half the maximum positive temperature gradient.

According to the statistical data about the temperature and the temperature gradient variation in Germany, studies have been conducted to analyze the slab stress state under the action of the temperature gradient, especially the slab with smaller lateral size whose warping deformation is not restrained completely. The calculation model with discontinuous supporting was utilized to calculate the warping stress [32] under the action of dead load and temperature gradient.

The warping stress and displacement of the slab track in different constraint conditions were analyzed by finite element theory. The results show that the stronger the restraint acting on the track slab, the more the warping deformation is resisted, and the closer the warping stress in the slab track to that of an infinite slab. The restraints acting on the track slab include the track dead load, the restraint of the continuous long rails, and the train load acting on the rails. Because of the large supporting coefficient in the ballastless track supporting system, the


loading restraint of the track slab, due to the limitation of the loading magnitude and position, shows many differences. For convenience, no matter for the unitary or the continuous structure, the warping stress of the ballastless track in the longitudinal or lateral direction is calculated in accordance with the infinite slab.

VI. CALCULATION OF FOUNDATION DEFORMATION EFFECT

Ballastless track will be influenced remarkably by the large rigidity of the track slab or the bed slab once uneven deformation occurs at the foundation.

In the Japanese slab track base design, the maximum settlement displacement (δ) occurs at the mid-point and at the ends of the baseplate with the half-wave sinusoid of δ=20 mm at the service and fatigue state, together with that of δ=30 mm at the ultimate state. Based on the deformation relevance, the rigidities at different locations of the settlement area with an interval of 5 m are calculated to ensure the settlement of the baseplate under the dead load reaches the designed uneven settlement. Then the additional bending moment [33] due to foundation deformation of the baseplate is calculated. Germany has a concept of “zero settlement” that uneven settlement must not occur. Thus, there is no need to consider the uneven settlement effect in design. Although high-speed railways have developed rapidly in China, uneven settlement is also inevitable at the subgrade-bridge transitional sections and high embankment. In order to ensure the proper operation of ballastless track, the influence of uneven settlement of foundation should be considered in the design of ballastless track in China.

Because of the large rigidity of ballastless track, when there is uneven settlement, the slab will have the same deformation as the foundation, which can be viewed as a forced displacement of the slab structure. In this case, the bending moment of the slab under the action of foundation deformation equals to the product of its flexural rigidity and the uneven deformation curvature.

VII. DESIGN OF BALLASTLESS TRACK STRUCTURE

The bearing structures of the ballastless track mainly include plain concrete, reinforced concrete, and prestressed reinforced concrete. The plain concrete structure is usually applied to the tunnels with good foundation and small ambient temperature variation. In this case, the slab will not crack [34] under the action of train load and environmental factors. Under the common foundation conditions, the slab may readily crack with the influences of foundation deformation, train load, and environmental factors. Thus, it is necessary to add reinforcements to limit the crack development. As for the continuous reinforcement concrete slab, because the temperature stress is the main influencing factor, reinforcements are laid near the neutral axis to limit the crack width and crack interval of the track slab. For the sections with severely weak foundations, the bending moment in the slab is usually large. In order to limit the crack width, we need to thicken the slab or improve the foundation, which results in high costs. In that case, placing reinforcements in top and bottom layers can help the track slab bear more bending


moment [35-36]. To limit the crack width and improve the structure durability, steel fiber concrete has been increasingly used in the ballastless track structure [37-38]. In cold areas, prestressed reinforced concrete structure is often adopted for decreasing the freezing injury.

The allowable stress method and the ultimate state method are generally utilized in the concrete structure design. As the Japanese track slab was designed as reinforced concrete structure originally, the allowable stress method is adopted provided that the track slab concrete under the action of bending moment conforms to the hypothesis of plane mechanism, while the tensile stress of the concrete in the tension zone is negligible. The allowable stress of the reinforcement depending on the repeated loading times varies with different design wheel load and structure types. In the cold areas, anti-freezing measures should be taken. Considering factors such as construction and costs, the prestressed reinforced concrete structure [39] designed by partial limit state theory is applied.

For the German ballastless tracks like Rheda and Züblin, the longitudinal reinforcements are placed in the continuous slab for the purpose of controlling the crack types and width. The width of the slab is determined by the Westergaard’s stress equations and the allowable compressive stress of the subgrade surface. Determination of the slab thickness follows the principle that the stress caused by temperature gradient and load is less than the flexural strength of the slab concrete. The supporting layer is composed of plain concrete probably with cracks or is the hydraulic supporting layer structure. The load stress should be checked within the permissible limit to determine the modulus of elasticity of the supporting layer.

Ballastless track, laid on the elastic foundation under the long-term repeated action of train load and environmental change, is of band structure distinct from the structures like bridge and building. In order to ensure its high accuracy and high stability, the rail structure is required to work in an elastic condition under the action of train load and surrounding factors. Therefore, we suggest that the ballastless track structure design adopts the allowable stress method for the innovative research of high speed railway in China.

During the design, it is assumed that every plane cross-section remains a plane under the action of the bending moment. The normal stress of the concrete in the compression zone takes a triangle pattern, the tensile strength of the concrete in the tension zone is neglected for the reinforced concrete components, and the normal stress of the concrete in the tension zone also takes a triangle pattern for prestressed reinforced concrete components. Under the action of axial force, the temperature stress of the continuous ballastless track, which may cause cracking, is resisted by the reinforcement, and the plane assumption is invalid.

Because the method for calculating the load effect, especially under the action of bending moment is different from the calculation model for structure design, a correction factor is introduced to eliminate the difference in the obtained results. The specific design flow for the ballastless track structure is shown in fig. 2.


As for the unit ballastless track, the reinforcement is mainly based on the load bending moment and the effect temperature force is negligible. For the continuous ballastless track, concrete contraction and temperature decreases are the main factors influencing the reinforcement. The load combinations for the different kinds of ballast-less track on diverse foundations are listed in table 1.

The daily temperature has a periodic variation, leading to a periodic variation in temperature stress and warping stress. Nevertheless, the maximum temperature gradient and the maximum temperature force do not appear everyday. Especially for the continuous ballastless track, the maximum temperature force only occurs at the critical state when a new crack appears. When the crack is stabilized, the temperature force is mostly less than the maximum temperature tension. Therefore, the maximum temperature tension is unlikely to appear in the continuous ballastless track, and it can be regarded as a kind of load combination and checked independently.

The subgrade of PDLs is required to be designed and constructed under the concept of “zero settlement”. However, uneven settlement is easy to occur at the transitional section between subgrade and the structures such as bridge, tunnel or culvert. And the probability of the uneven settlement within a small range occurring to common sections is quite low. Therefore, the uneven settlement of subgrade should be combined as an additional force with a low probability of occurrence. Under the train load, a bridge has bending deformation which coincides with the train load. Consequently, the bridge bending deformation should be combined as the main force the same as the train load.

Fig 2. Design flow for the ballastless track structure

Tentative structure size

Tentative reinforcement

Internal force calculation

Train load bending moment

Temperature gradient

Foundation deformation bending moment

Temperature force

Load stress σ

Reinforcement optimization

Finish design

Allowable stress [σ]

Structure functional design

Train load bending moment structure

coefficient

Temperature gradient bending moment

structure coefficient

Foundation deformation bending moment

structure coefficient

Structure coefficient calculation

σ<[σ]


Table 1. Suggested load combinations for different types of ballastless tracks

Load combination Type

On subgrade On bridge In tunnel

3MTrain 3MTrain 3MTrainMain force

1.5MTrain+M∆T 1.5MTrain+M∆T+ MD - Unit

ballastless track Main force

+additional force 1.5MTrain+M∆T+ MD - -

3MTrain 3MTrain 3MTrain

FT, max FT, max FT, maxMain force

1.5MTrain+M∆T+FT 1.5MTrain+M∆T+ MD +FT+ FD 1.5MTrain +FT

Continuous ballastless

track Main force

+additional force 1.5MTrain+M∆T+MD+FT 1.5MTrain+M∆T+ MD +FT+FB+FB D -

Note: MTrain denotes the bending moment caused by train load, M∆T denotes the bending moment caused by temperature gradient, MD denotes the bending moment caused by foundation deformation, FT,max denotes the maximum temperature force, FT denotes the temperature force, FD denotes the axial force caused by foundation deformation, and FB denotes the braking force.

According to the load combinations shown in Table 1, one should decide whether the edge stress of the supporting layer in the ballastless track exceeds its cracking stress. If the edge stress is lower than the cracking stress, then the supporting layer of the concrete will not crack, and reinforcement is unnecessary or should be placed in accordance with the structure. If the edge stress is higher than the cracking stress, then the supporting layer will crack. Especially for the continuous ballastless track, full-section cracking is likely to occur. At the moment, all the concrete in the tension zone at the cracking location under the action of bending moment stops working, and all the tension is resisted by the reinforcement. Nevertheless, the concrete between two cracks in the tension zone is still functioning, which leads to the variation of sectional flexural rigidity and neutral axial. The flexural rigidity of the cracked slab is decreased sharply. A thinner slab with a higher concrete grade will have a smaller flexural rigidity after cracked.

The flexural rigidity used for the load stress calculation is the one with the supporting layer’s full section sharing the stress. However, during the design the concrete in the tension zone is supposedly out of operation completely, from which some error will occur and there is a need for revision.

According to the elastic foundation beam theory, the bending moment of the foundation beam under the concentrated load (train load) is directly related to the coefficient of elasticity of the foundation and the flexural rigidity of the foundation beam. The bending moment of the foundation beam is directly proportional to the 1/4 power of its flexural rigidity. Similarly, the bending moments of the slab caused by temperature gradient and foundation deformation are


directly proportional to its flexural rigidity. Thus, the correction factor of the bending moment caused by train load, temperature gradient and foundation deformation can be obtained.

VIII. CONCLUSIONS AND SUGGESTIONS

Ballastless track, with the merits of good ride comfort, high stability and little maintenance, has become the main type of the rail structure. The design concepts of the ballastless track are different in different countries: the factors considered in design and the calculation methods vary greatly with each other. This paper has summarized and analyzed calculation and design methods for ballastless track in the world. Based on the re-innovation research results of the ballastless track in China, relatively general design concepts and methods for ballastless track were put forward tentatively, which guided the design of ballastless track on the Suining-Chongqing test section, the Wuhan-Guangzhou passenger dedicated line, the Lanzhou-Urumchi No.2 double line, as well as the reference diagram design of the slab track and the double block track.

Although, a type of structure with little maintenance, the ballastless track has many conditions during operation. Therefore, the design theory of ballastless track still needs further study. The future work may involve the following:

(1) Research on the fatigue properties under the coupling action of train and temperature load. Train load and temperature load are two kinds of loads repeatedly acting on the ballastless track. The statistical characteristics of train load and temperature load, especially the fatigue properties under different loads and their coupling action, should be studied to provide a basis for predicting the fatigue life of ballastless track.

(2) Research on the durability of ballastless track. The ballastless track is a composite structure composed of many kinds of materials. Under the combined action of environment and train load, different kinds of materials have different degradation curves, and the damage in one component will influence the durability of the whole structure. Therefore, a systematic method should be applied to the durability research of ballastless track, to realize a design concept of little maintenance.

(3) Research on long-term dynamic properties. Under the long-term combined action of train load induced vibration and natural environment, the function of the components of ballastless track is likely to degrade gradually. Consequently, the dynamic characteristics of ballastless track, and the safety and stability of train will be influenced. An analysis model of ballastless track with damage should be established for research on the long-term dynamic properties of ballastless track.

(4) Research on maintenance mechanics. At the beginning of the construction of ballastless track, some conditions have already occurred because of the errors in design and construction. Thus far, there is lack of a systematic, intensive study on the causes of diseases and the


countermeasures. The intensive study on disease mechanism, maintenance standard, maintenance time, maintenance method, and the influence of maintenance on track and train will lay a good foundation for the maintenance work of ballastless track.

ACKNOWLEDGEMENTS

This work was supported by the National Natural Science Foundation of China (No. 51008258), and the Fundamental Research Funds for the Central Universities (No. SWJTU09BR038).

References [1]. Y.X. Wei, Y.J. Qiu, Research on the subgrade surface stiffness of ballastless track of high-speed railway, Journal of Railway Engineering Society, 2010(7): 15-20 (in Chinese). [2]. L. Bernhard, Developments in road pavement construction and railway track technology for a sustainable surface transport infrastructure, In: The Emerging Frontiers of Transportation and Development in China, Chengdu, 2009. [3]. D.Z. Zheng, D.C. Feng, Mechanics of Layered Elastic System, Harbin: Harbin Institute of Technology Press, 2001: 126-133 (in Chinese). [4]. Ando Satoshi, Loading-deformation characteristics of slab track with temperature variation, Railway Summary Report, 1989, 10(3): 2-9. [5]. Q.C. Wan, Y.G. Lu, Stress Calculation of New Sub-rail Foundations, Beijing: China Railway Publishing House, 1987: 15-45 (in Chinese). [6]. C. Esveld, Modern Railway Track, 2nd ed., Zaltbommel: MRT Productions, 2001: 71-73. [7]. Y.X. Liu, X.F. Chen, Two analytical methods of slab track structures. Urban Mass Transit2007(6)32-34,66 (in Chinese). [8]. C.Y. Qi, Research of soil subgrade slab track structures strength calculation, Railway Standard Design, 2006(2): 26-28 (in Chinese). [9]. L. Sun, Double-block ballastless track design of Wu-Guang passenger railway, Railway Standard Design, 2006(Sup.): 155-158 (in Chinese) [10]. S. Yoshihiko, New Track Mechanics, trans. Y. Xu, Beijing: China Railway Publishing House, 2001: 238-240 (in Chinese). [11]. J. Eisenmann, G. Leykauf, Slab track for railways, In: Concrete Calendar 2000, Berlin: Verlag Ernst & Sohn, 2000: 10-15 (in German). [12]. J.J. Fan, Modern Railway Track, Beijing: China Railway Publishing House2000: 43-51 (in Chinese). [13]. J. Eisenmann, The dimension of a type of slab track, Railway Engineering, 1991, 42(3): 116-118, 120-122, 124 (in German). [14]. Q.D. YueF.X. Jiang, X.G. Li, A study on mechanical model of two-level crossing superposed beam system for section track structure, Railway Standard Design, 2002(12): 1-4 (in Chinese).[15]. S.Q. Li, S.J. Duan, J.W. Wang, et al., Track structure model of elastic supported grillage girder, Engineering Mechanics, 2001(Sup.): 508-513 (in Chinese). [16]. J.W. Wang, S.J. Duan, S.Q. Li, High-speed railway slab track structure static analysis, Engineering


http://dlib.edu.cnki.net/KNS50/Navi/Bridge.aspx?LinkType=BaseLink&DBCode=cjfd&TableName=cjfdbaseinfo&Field=BaseID&Value=TDGC&NaviLink=%e9%93%81%e9%81%93%e5%b7%a5%e7%a8%8b%e5%ad%a6%e6%8a%a5

Mechanics, 2002(Sup.): 290-293 (in Chinese). [17]. C.X. Liu, W.M. Zhai, Elementary research on problem of slab’s intensity by means of finite element analysis, Journal of Railway Engineering Society, 2001(1): 24-26 (in Chinese). [18]. L. Gao, M.N. Ma, D.M. Wang, Mechanic characteristic research of ballastless track structures on bridge with linear motion actuator delivery system, Railway Standard Design, 2007(7): 5-7 (in Chinese). [19]. C.X. Li, C.H. Li, Z.H. Kou, Mechanics analysis of slab track on soil foundation, Railway Engineering, 2005(7): 88-90 (in Chinese). [20]. P.R. Zhao, X.Y. Liu, Dynamic characteristic analysis and parameter study of slab track, Railway Engineering, 2004(5): 48-50 (in Chinese). [21]. P.R. Zhao, Y.A. Zhang, X.Y. Liu, et al., Beam-plate model on the elastic foundation of ballastless track, China Railway Science, 2009(3): 1-4 (in Chinese). [22]. J. Ren, Preliminary research and design of slab track, Railway Standard Design, 1996(6): 25-28 (in Chinese). [23]. Q.C. Wang, Design and Construction of Slab Track, Chengdu: Southwest Jiaotong University Press, 2002: 52-56 (in Chinese). [24]. C.B. Cai, P. Xu, Dynamic analysis of key design parameters for ballastless track of high-speed railway, Journal of Southwest Jiaotong University, 2010, 45(4): 493-497 (in Chinese). [25]. P. Xu, C.B. Cai, Theoretical calculation of the supporting stiffness of the subgrade surface in ballastless track and its application, China Railway Science, 2010(1): 21-25 (in Chinese). [26]. P.R. ZHAO, X.Y. LIU, Reduced elastic modulus of cracked support layer in twin-block ballastless track, Journal of Southwest Jiaotong University, 2008, 43(4): 459-464 (in Chinese). [27]. Rail. ONE GnmH, Rheda 2000 Crack width calculation according to DIN 1045-1, Ingolstaedter, 2005. [28]. P.R. Zhao, X.Y. Liu, Thermal stress calculation method research of continuous slab, Railway Standard Design, 2008(10): 6-8 (in Chinese). [29]. P.R. ZHAO, X.Y. LIU, Thermal stress calculation method and parameter analysis of continuous slab, Railway Engineering, 2008(11): 81-85 (in Chinese). [30]. S.R. Wang, Study on Temperature Stress for Ballastless Track Slab [Dissertation]. Chengdu: Southwest Jiaotong University, 2007 (in Chinese). [31]. R.S. Yang, Experimental report of track static test on Jialing river bridge, Chengdu: Southwest Jiaotong University, 2007: 21-25 (in Chinese). [32]. J. Eisenmann, G. Leykauf, Concrete Carriageways, Munchen: Ernst&Sohn, 2003: 22-59 (in German). [33]. Japan Railway Construction Public Corporation Morioka Branch Office, Design Calculation of RC Subgrade En-gineering, t=300 with Detail Design of Northeast Trunk Line, RC Subgrade Engineering, Japan, 1999. [33]. A. Faeh, M. Gloor, T. Gerber, Optimised ballastless track systems, In: IABSE Symposium, Antwerp, Belgium, August 2003: 52-60. [34]. C. Esveld, V. Markine, Use of expanded polystyrene (EPS) sub-base in railway track design, In: IABSE Symposium, Antwerp, Belgium, August 2003: 14-15. [35]. J.M. Zwarthoed, V.L. Markine, C. Esveld, Slab track design: flexural stiffness versus soil improvement, In: Proceedings of Rail-Tech Europe 2001 Conference, Utrecht, April 2001: 1-22. [36]. V. Henke, H. Falkner, SFRC for jointless railway tracks, In: IABSE Symposium, Antwerp, Belgium, August 2003: 46-51. [37]. T. Takahashi, F. Sekine, T. Horiike, et al., Study on the applicability of short fiber reinforced concrete to precase concrete slabs for slab track, Quarterly Report of RTRI, 2008, 49(1): 40-46. [38]. Japan Railway Construction Public Corporation Morioka Branch Office, Design Calculation of Northeast Trunk Line, Slab Design of PRC Structure A-55C, Japan, 1999♦


http://dlib.edu.cnki.net/KNS50/Navi/Bridge.aspx?LinkType=BaseLink&DBCode=cjfd&TableName=cjfdbaseinfo&Field=BaseID&Value=XNJT&NaviLink=%e8%a5%bf%e5%8d%97%e4%ba%a4%e9%80%9a%e5%a4%a7%e5%ad%a6%e5%ad%a6%e6%8a%a5

http://dlib.edu.cnki.net/KNS50/Navi/Bridge.aspx?LinkType=BaseLink&DBCode=cjfd&TableName=cjfdbaseinfo&Field=BaseID&Value=ZGTK&NaviLink=%e4%b8%ad%e5%9b%bd%e9%93%81%e9%81%93%e7%a7%91%e5%ad%a6

http://dlib.edu.cnki.net/KNS50/Navi/Bridge.aspx?LinkType=BaseLink&DBCode=cjfd&TableName=cjfdbaseinfo&Field=BaseID&Value=XNJT&NaviLink=%e8%a5%bf%e5%8d%97%e4%ba%a4%e9%80%9a%e5%a4%a7%e5%ad%a6%e5%ad%a6%e6%8a%a5

AN EFFICIENT METHOD FOR VIETNAM LICENSE PLATE LOCATION

MAI VINH DU DUOQIAN MIAO RUIZHI WANG Department of Computer Science and Technology Tongji University, Shanghai 201804, China LE HUNG LAN Electrical & Electronic Department, The University of Transport and Communications, Hanoi, Vietnam

Summary: In the Automatic License Plate Recognition (ALPR) system, the License Plate Location (LPL) is the key step before the final recognition. This paper proposed a LPL algorithm for Vietnam license plates, which combined pre-processing, morphology open & dilation operation, measure properties of image regions to find candidates and then finding the license plate angle & rotating license plate. The proposed algorithm consists of three main modules: Pre-processing (convert RGB image to grayscale image, image binarization use Otsu method), License plate candidates location (morphology open to remove noises & dilation operation, measure properties of image regions to find candidates), License plate exactly location (finding the license plate angle & rotating license plate, cut exactly license plate region). Experiments have been implemented on 350 Vietnam license plate images taken from various scenes, including diverse angles, different lightening conditions (night and day), reflected light and the dynamic conditions. The efficiency of processing of the proposed algorithm is improved and average rate of accuracy of the license plate locating is 98.64%.

Keywords: License plate location, mathematic morphology, image converting, Ostu method, image rotating, image measuring.

I. INTRODUCTION

Automatic License Plate Recognition (ALPR) is an important part of Intelligent Transportation System (ITS) and can be utilized in many ITS applications such as: traffic management system, traffic information system, traffic volume control system, none-stop toll collection system & one-stop toll collection system, information traveling system, weigh in motion system, car-park system, and so on. Usually, an ALPR system consists of three parts: 1) license plate location, 2) character segmentation, 3) character recognition. Among these three parts, the most important and basic is to license plate location step, which directly affects system’s overall accuracy. There are many techniques have been proposed to address vehicle


license plate location since the 90’s decade, and a number of techniques have been proposed, such as the methods base on histogram and mathematical morphology [1-9], wavelets transform [10], texture feature analysis [11], edge detection analysis [12], color feature [13], region-based segmentation [14]. But most previous works have in some way restricted their working conditions, such as limiting them to indoor scenes, fixed illumination, prescribed driveways, and these methods only implemented for their image data base, so it may be not good for Vietnam license plate.

In this paper, we had considered for the specific characteristics of Vietnam license plates, we had implemented for the vehicle images, which obtained from the actual system, these vehicle images are very different background, different illumination, different license angles, different sizes and types of license plates, different dimensions from camera to vehicles, different light conditions in Vietnam environment. This paper put forward a method for the Vietnam license plate location, the proposed algorithm consists of three main modules: Pre-processing (convert RGB image to grayscale image, image binarization use Otsu method), License plate candidates location (morphology open & dilation operation, measure properties of image regions to find candidates), License plate exactly location (finding the license plate angle & rotating license plate, cut exactly license plate region). The rest of the paper is organized as follows: section 2 introduce Feature of Vietnam license plates, section 3 shows The Algorithm of Vietnam license plate location, section 4 shows the experiment results and section 5 are the conclusions and lasts are acknowledements and the references.

II. FEATURE OF VIETNAM LICENSE PLATES

Based on the "Circular 06/2009/TT-BCA-C11 regulations on vehicle registration issued by the Ministry of Public Security" issued on 11/03/2009, according to that, if we divide types of license plate in color, the license plates (LP) of Vietnam have four types: Type 1 - the license plate with white background and black characters & numerals, Type 2 - the license plate with blue background and white characters & numerals, Type 3 - the license plate with red background and white characters & numerals, Type 4 - the license plate with the yellow background and black characters & numerals. In this paper, we only consider to locate for Type 1, this category has large number, over 90% of vehicles in Vietnam, which the license plate with white background and black characters & numerals. Numerals are from 0 to 9 and characters use a serial symbol is one of the 15 digits F, H, K, L, M, N, P, R, S, T, U , V, X, Y, Z. This type also divided in two subtype (one-row LP and two-row LP), the shaps and dimensions of this type such as table 1.

Table 1. Shaps and dimensions of Vietnam license plates of Type 1 Type Shape Dimension

One-row LP

Two-row LP


III. THE ALGORITHM FOR VIETNAM LICENSE PLATE LOCATION

3.1. Flowchart of the proposed algorithm

This paper combined pre-processing, morphological operations (open operation, dilate operation) [1-9], measure properties of image regions to find candidates, last is cut exactly license plate region, the flowchart such as figure 1.

Fig 1. Flowchart of the proposed algorithm for Vietnam License Plate Location

3.2. Pre - Processing

The performance of the ALPR system is heavily dependent on the environmental set-up, such as the quality of cameras, light conditions, and other forms of controlling the environment. Since the images of license plate are acquired from the various complicated background, e.g. contrast, illumination, blurring, dirty, sizes and weather conditions, it is difficult for LPL work especially when the acquired images are with thick shades and in low contrast or bad quality [7, 11, 12], etc. So the vehicle images pre-processing before LPL work in the ALPR system is necessary.

3.2.1. Convert RGB image to Grayscale image

First one must obtain the values of its red, green, and blue (RGB) primaries in linear intensity encoding, by gamma expansion. Then, add together 30% of the Red value, 59% of the Green value, and 11% of the Blue value (these weights depend on the exact choice of the RGB primaries). We converts RGB values to grayscale values by forming a weighted sum of the R, G, and B components, with 8 bits per pixel, which allows 256 different intensities. The effective


http://en.wikipedia.org/wiki/RGB

http://en.wikipedia.org/wiki/Gamma_correction

luminance of a pixel is calculated by the formula 1. Figure 2(a) is origin RGB image, the output grayscale image is figure 2(b).

(1)

(a)

(b)

Fig 2. a) Origin RGB image; b) Grayscale image

3.2.2. Adjust grayscale image intensity

We should adjust old intensity values of grayscale image to new values, the image is weighted toward lower (darker) output values. Intensity adjustment is image enhancement techniques are used to improve an image, where "improve" is sometimes defined objectively, and sometimes subjectively. For intensity images, the n bins of the histogram are each half-open intervals of width A/(n-1), the pth bin is the half-open interval.

(2)

Where x is the intensity value. The scale factor A depends on the image class, A is 255 if the intensity image of class uint8. Figure 3(a) is histogram of grayscale image, the figure 3(b) is histogram of adjusted grayscale image intensity.

(a)

(b)

Fig 3. a) Histogram of grayscale image; b) Histogram of adjusted grayscale image intensity

3.2.3. Image binarization using Otsu Method

Now, we convert adjusted image grayscale image in figure 4(a) to binary image based on threshold, we threshold a gray-level image to binary image, a simple but effective tool to separate objects from background flowing formula.

(3)

Where, T is some global threshold. Global image threshold using Otsu's method, computing a global threshold (level) that can be used to convert an intensity image to a binary image, level is a normalized intensity value that lies in the range [0, 1], which chooses the threshold to minimize the intraclass variance of the black and white pixels. Otsu Method-


Principle - Algorithm as following:

• Define the within-class variance:

(4)

Where

[0, L-1] the range of intensity level, is the variance of the pixel in the background (below threshold, is the variance of the pixel in the foreground, above threshold).

• Easier for the between-class variance:

(5)

By simplifying

(6)

Where

• Easier using simple recurrence relations

(7)

(8)

(9)

(10)

From σ2 and μ, we define optimal threshold T, and then we convert to binary image following formula (3). Figure 4(b) is image after convert to binary image, based on threshold (Otsu's method).

(a)

(b)

Fig 4. a) Adjusted image grayscale image, b) Binary image based on threshold (Otsu's method)

3.3. License Plate Candidates Location

3.3.1. Morphological opening operation

The opening serves in image processing as a basic workhorse of morphological noise removal. Morphological opening [3] used to remove small objects (connected component) from the foreground (usually taken as the dark pixels) of an image, placing them in the background.


http://en.wikipedia.org/wiki/Image_processing

The opening of A by B is obtained by the erosion of A by B, followed by dilation of the resulting image by B.

(11) Where and denote erosion and dilation. We use this method for binary image to

improve the binary image, we removes objects binary image based only on their pixel area from a binary image all connected components (objects) that have fewer than 2000 pixels (removed noises, which are not license candidates), producing another binary image. The basic steps of this method are 1) determine the connected components, 2) compute the area of each component, 3) remove small objects. After implement we have figure 5(a).

(a)

(b)

Fig 5. a) After morphologically opening to remove noises, b) After morphologically dilation by a SE

3.3.2. Morphology dilation operation

Dilation operation used to increase objects in the image, definition for binary images, the positions where a given structure element fits. The dilation of A by the structuring element (SE) B is defined by.

(12)

Dilate is a function also known by the names "grow", "bolden", and "expand". It turns on pixels which were near pixels that were on originally, thereby thickening the items in the image. Algorithm of dilate operation is automatically takes advantage of the decomposition of a structuring element object (if it exists), also when performing binary dilation with a structuring element object that has a decomposition, it is automatically uses binary image packing to speed up the dilation. Figure 5(b) is image after dilate operation.

3.3.3. Measure properties of cadidate regions (connected components)

From binary image obtained in the figure 5(b), we will locate license plate candidate by two steps as following.

Step 1 - Convert numeric values to logical: After convert figure 4 (b) to logical values, we have a logical image where is a logical array that is the same size as figure 5(a).

Step 2 - Measure properties of image regions (connected components): We measures a set of properties for each connected component (object) in the binary image (the image is a logical array). The fields of the structure array denote different properties for each region, properties


http://en.wikipedia.org/wiki/Opening_(morphology)

http://en.wikipedia.org/wiki/Dilation_(morphology)

can be a comma-separated list of strings, a cell array containing strings, the actual number of pixels in the region measured by shape measurements. The smallest rectangle containing the region, a 1-by-Q *2 vector, where Q is the number of image dimensions, where is in the form [x y] and specifies the upper-left corner of the bounding box, is in the form [x width y width] and specifies the width of the smallest rectangle containing the region along each dimension. Selecting the license plates from the candidates, we chosen area is the deepest region in the frame which has properties of the license plate area (min area of LP and max area of LP, max ratio and min ratio between height and width of LP, max height and min height, max width and min width). And then, finding the components which are at the depth "depth", set the coordinates of the supposed license plate region. After that a license plate candidate is cut from theses coordinates such as figure 6.

Fig 6. License plate cadidate obtained by measure properties of cadidate regions

3.4. License plate exactly location

3.4.1. Finding the LP angle & rotating LP In fact, the license plates usually not perpendicular in the images, so we need determine the

angle of the plate, and then if detected an angle we will rotating LP. Finding the LP angle: The image contains parallel lines or at least one large line. Applying

the Radon transform on an image f(x,y) for a given set of angles can be thought of as computing the projection of the image along the given angles. The result is a new image R(θ,ρ), this is depicted in figure 7. This can be written mathematically by defining (13)

After that, the Radon transform can be written as:

(14)

Where δ(.) is the Dirac delta function. There are two distinct Radon transforms (the source can either be a single point or it can be a array of sources). In case a array of sources, the source and sensor contrapment is rotated about the center of the object. For each angle, the density of the matter the rays from the source passes through is accumulated at the sensor. This is repeated for a given set of angels, usually from ε [0,180]. The angel 180 is not included since the result would be identical to the angel 0, as figure 7. The Radon transform is a mapping from the Cartesian rectangular coordinates (x,y) to a distance, also known as polar coordinates. Radon transform algorithm includes some step as following: The Radon transform of an image is the sum of the Radon transforms of each individual pixel. The algorithm first divides pixels in the image into four subpixels and projects each subpixel separately. Each subpixel's contribution is proportionally split into the two nearest bins, according to the distance between the projected location and the bin centers. If the subpixel projection hits the center point of a bin, the bin on


the axes gets the full value of the subpixel, or one-fourth the value of the pixel. If the subpixel projection hits the border between two bins, the subpixel value is split evenly between the bins.

Fig 7. Radon transform to find LP angle

After finding the LP angle, we have a image with lines as figure 8(a).

(a)

(b)

Fig 8. a) Image after found angle of LP, b) Image after rotating

Rotating LP: Since angle of license plate determined by Radon transform, we can implement rotating this image, using the interpolation method specified by bilinear interpolation, after rotating, we obtained figure 8(b).

3.4.2. Cut exactly license plate region

We find the coordinates of the LP rectangle inside the image, find contours for the sum of the lines and of the columns in the image. Find contours are find the x-coordinates respectively of the first point at the left and the first point at the right of the vector which is superior or equal to the average of the vector. This permits to delimit the plate eliminating noises around it. We rotate the image using the interpolation method specified by bilinear interpolation. We have a four-element position vector [xmin ymin width height] that specifies the size and position of the crop rectangle. After cut exactly license plate region, we have the result as figure 9.

Fig 9. Exactly license plate region

IV. EXPERIMENT RESULTS

We implemented experiment with PC Intel(R) Core(TM)2 Duo CPU T7250 @2.00GHz, RAM 1.00 GB, Windows Vista Version 6.1 (Build 7600) 32-bit Operating System and MATLAB Version 7.8.0.347 (R2009a). We implemented test for 350 vehicle images, which obtained from the actual system, these vehicle images are very different background, different illumination, different license angles, different size and type of license plates, different dimensions from camera to vehicles, different light conditions in Vietnam environment. There


are two type of the size of the images in RGB true-color image was tested (800x600 pixels and 768x288 pixels). The results of the our method and the previous works show in table 2. The average rate of accuracy is 98.64%, comparision with previous methods, our results are more axactly, some examples of the experiments show in figure 10.

Table 2. Experiment results of our method and previous works Experiment results of our method Previous works

Type of LP

Total image

Rate of accuracy

Average rate of

accuracy

Accuracy of [2]

Accuracy of [3]

Accuracy of [4]

Accuracy of [6]

One-row LP 200 98.97%

Two-row LP 150 98.30%

98.64% 96.50% 97.00% 96.00% 83.50%

Fig 10. Some examples of our experiment results

V. CONCLUSIONS

This paper put forward a LPL algorithm for Vietnam license plates, the proposed algorithm consists of three main modules: Pre-processing (convert RGB image to grayscale image, image binarization use Otsu method), License plate candidates location (morphology open & dilation operation, measure properties of image regions to find candidates), License plate exactly location (finding the license plate angle & rotating license plate, cut exactly license plate region). Experiments have been conducted for this algorithm, which implemented test for 350 vehicle images, which obtained from the actual system, these vehicle images are very different background, different illumination, different license angles, different size and type of license plates, different dimensions from camera to vehicles, different light conditions in Vietnam environment. The algorithm can quickly and correctly detect the region of license plate and the


average rate of accuracy is 98.64%, comparision with previous methods, our results are more axactly. From the result of the experiment, we can see the proposed approach is robust. But there are still some images failed in the experiment, our algorithm still needs further research.

VI. ACKNOWLEDEMENTS

This work was supported by National Natural Science Foundation of China (Serial No. 60970061, 61075056). And supported by Research team of Electrical & Electronic Department, The University of Transport and Communications, Hanoi, Vietnam.

References [1]. Dubey, P., “Heuristic approach for license plate detection,” in Annual IEEE INDICON, pp. 366-370, 2005. [2]. Mahini, H.; Kasaei, S.; Dorri, F., “An Efficient Features - Based License Plate Localization Method,” in 18th International Conference on Pattern Recognition, vol. 2, pp. 841-844, 2006. [3]. W. K. I. L Wanniarachchi, D. U. J. Sonnadara and M. K. Jayananda, “Detection of License Plates of Vehicles,” in Proceedings of the Technical Sessions, vol. 23, pp. 13-18, 2007. [4]. Humayun K. Sulehria, Ye Zhang, Danish Irfan, “Mathematical Morphology Methodology for Extraction of Vehicle Number Plates,” in INTERNATIONAL JOURNAL OF COMPUTERS , vol. 1, pp. 69-73, 2007. [5]. Gisu Heo, Minwoo Kim, Insook Jung, Duk-Ryong Lee, II-Seok Oh, “Extraction of Car License Plate Regions Using Line Grouping and Edge Density Methods,” in International Symposium on Information Technology Convergence, vol. , pp. 37-42, 2007. [6]. Faradji, F., Rezaie, A.H., Ziaratban, M., “A Morphological-Based License Plate Location,” in IEEE International Conference on Image Processing, vol. 1, pp. 57-60, 2007. [7]. Haiqi Huang, Ming Gu, Hongyang Chao, “An Efficient Method of License Plate Location in Natural-Scene Image,” Fifth International Conference on Fuzzy Systems and Knowledge Discovery, vol. 4, pp. 15-19, 2008. [8]. Hui Li, Pan Gao, Dongxiu Wang, Wangming Chen, “Study on Mathematical morphology based vehicle license plate location,” in 2nd International Conference on Power Electronics and Intelligent Transportation System (PEITS), vol. 2, pp. 116-119, 2009. [9]. Wei Pan, Rong An, “Morphology and auto -correlation based method of fast locating vehicle license plate,” in 2nd International Conference on Advanced Computer Control (ICACC), vol. 3, pp. 116-119, 2010. [10]. Feng Yang; Fan Yang, “Detecting license plate based on top-hat transform and wavelet transform,” in International Conference on Audio, Language and Image Processing, vol. , pp. 998-1003, 2008. [11]. Jun Kong; Xinyue Liu; Yinghua Lu; Xiaofeng Zhou, “A novel license plate localization method based on textural feature analysis,” in Fifth IEEE International Symposium on Signal Processing and Information Technology, vol., pp. 275-279, 2005. [12]. Xing Yang; Chaochao Huang; Hua Yang;, “Research on Adaptive Preprocessing License Plate Location,” in The 9th International Conference for Young Computer Scientists, vol. , pp. 764-768, 2008. [13]. Xiaoqian Liu; Weiqiang Wang; Tingshao Zhu;, “A Method Based on Character Edge Color for Quick Locating Vehicle License Plate,” in 20th International Conference on Pattern Recognition (ICPR), vol. , pp. 3232-3235, 2009. [14]. Wenjing Jia, Huaifeng Zhang, Xiangjian He, “Region-based license plate detection,” in Journal of Network and Computer Applications, vol.30 , pp. 1324-1333, 2006♦


http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=9258














ANALYSIS ON SETTING AIRSHAFT AT MID-TUNNEL TO REDUCE TRANSIENT PRESSURE VARIATION

YINGXUE WANG BO GAO CHAO ZHANG XUZHOU HE School of Civil engineering, Southwest Jiaotong University, Chengdu 610031, China

Summary: Setting airshaft is one of the efficient methods to control the aerodynamic effects during the train traveling through a tunnel. Using numeral simulation method, this paper analyzed the difference between setting airshaft and without setting airshaft in transient pressure in cabin. After setting airshaft, the strength of pressure fluctuation in cabin can be reduced more than 40%. By analyzing the rule of airshaft alleviating pressure fluctuation, the expression to determine the optimal airshaft position was deduced.

Key words: Tunnel; airshaft; high-speed train; transient pressure.

I. INTRODUCTION

When a high-speed train passes through a tunnel, compression wave and expansion wave will be generated and transmitted to and fro, resulting in micro-compression wave radiation at tunnel exit and pressure fluctuation in cabin [1]-[3]. To deal with this, considerable efforts [4]-[8] have been made to alleviate the aerodynamic effects in the process of railway passing through tunnel, such as improving cabin’s airproof parameter, enlarging the cross section of tunnel, building hood at tunnel entrance. A desirable structure for a long tunnel is in the form of two shorter tunnels connected by pressure relief ducts [9]. Besides, airshaft is generally installed to reduce pressure intensity [10]-[11].

To controlling aerodynamic effects, most studies [5], [6], [8] and [12] focused on micro-compression wave at tunnel exit. In fact, the pressure fluctuation in cabin will reduce the comfort and even jeopardize passengers’ health; thus ameliorating transient pressure in cabin is an important project.

When a high speed train enters into a tunnel, the compression wave is generated ahead of the train due to the piston-like action of tunnel entry motion. The compression wave is discharged to the atmosphere as the form of a pulse-like wave, which is called the micro-pressure wave. The compression wave also arouses pressure fluctuation in cabin, which is called transient pressure. To relieve the intensity of the booming noise, many methods have been put forward, such as the installation of tunnel entrance hood, side tunnel branches, and shelters with slits linking adjacent tunnels [13]. Sealing carriage can realize the goal of reducing transient pressure and promoting passenger comfort. In practice, however, complete sealing of a train body is not possible because a train should have a lot of air conditioning devices, ventilation systems, etc. Thus, the large amplitude pressure variations on the train body during traveling


inside tunnel can penetrate into passenger’s room, often leading to ear discomfort in passengers [14[–[16]. This phenomenon will be more serious with the increase in train speed. The passenger’s ear discomfort inside train is, in general, associated with the magnitude of pressure variation, the rate of pressure variation, the sign of pressure variation, etc. [17]–[19]. In real high-speed railway trains, a special ventilation system is usually adopted to alleviate the pressure variations. This system controls the flow rate supplied into and exhausted from train, using a fan blower, according to the pressure variations occurring inside train. These pressure variations are proportional to the square of train speed [18]-[19]. The ventilation system may not be enough to reduce the pressure variations as the train speed increases. In order to alleviate the pressure variation inside train, several control methods have been investigated using a damper system, a continuous ventilation system, and a continuous ventilation control system [18-19]. All of these methods are to control the magnitude and the rate of the pressure variation occurring inside train.

Apart from solving pressure variation from the train point of view, building subsidiary construction in tunnel can also realize the goal. In this paper, using numeral simulation method, the compression wave transportation process in tunnel will be analyzed, and the transient pressure variation in cabin in the conditions of tunnel with airshaft and without airshaft will be compared.

II. GOVERNING EQUATIONS

During the course of a high-speed train passing through tunnel, the boundary condition is changed with time. The dynamic mesh model is applied to simulate this process.

The integral form of the conservation equation for a general scalar, φ, on an arbitrary control volume, V, with moving boundaries is written as [20]

d d ( ) dd

d d ,

gV V

V V

Vt

S Vφ

ρ φ ρ φ

Γ ∇ φ

∂

∂

+ − ⋅

⋅ +

∫ ∫

∫ ∫

u u A

A

=

(1)

Where ρ is the fluid density; is the flow velocity vector; u gu is the grid velocity of the

moving mesh; Γ is the diffusion coefficient; Sφ is the source term of φ; V∂ is used to represent the boundary of the control volume V.

Using a first-order backward difference formula, the time derivative term in (1) is written as:

1d ( ) (d ,d

n n

V

V VVt

ρφ ρφρφ+ −

=Δ∫

)t (2)

Where n and n+1 denote the respective quantity at the current and next time level. The (n+1)th time level volume Vn+1 is computed from

1 d ,d

n n VV Vt

+ t= + Δ

Where dV/dt is the volume time derivative of the control volume. In order to satisfy the grid conservation law, the volume time derivative of the control volume is computed from


,

d d ,d

fn

g g jVj

Vt ∂ j= ⋅ = ⋅∑∫ u A u A

(3)

Where nf is the number of faces on the control volume and is the j face area vector. The

dot product

jA

,g j ⋅u Aj on each control volume face is calculated by

, ,j

g j j

Vt

δ⋅ =

Δu A

(4)

Where jVδ is the volume swept out by the control volume face j over the time step . tΔ

Here, the large eddy simulation model (LES) is chosen. To model a free-stream condition at sites far from tunnel entrance and exit, pressure-far-field boundary conditions are adopted. The transmission medium is assumed as ideal gas.

III. AIRSHAFT EFFECTS

3.1. Calculation parameters The parameters of the railway, tunnel and airshaft are shown in table 1. To analyze the

effect of the shaft area on peak pressure at train body, the calculation results under different shaft open ratios will be given. The grid mesh at the tunnel entrance and train head is shown in fig 1. 3.2. Calculation results

Without airshaft, the transportations of compression wave and expansion wave and pressure fluctuation at train body are shown as fig. 2 and fig 3.

After setting shaft, the compression wave, expansion transportation process and pressure fluctuation on railway body are shown as fig. 4 and fig 5.

Table 1. Calculation parameter Carriage parameter Shaft parameter

Condition Tunnel length LT /m

Tunnel area AT /m2 Speed

V (km/h)Area (m2) Length (m) Shaft position Open ratio

Ⅰ — —

ⅠⅠ 14%

ⅠⅠⅠ

700 65 300 12.7 50 200 m to tunnel entrance 25%

Note: Shaft open ratio=shaft open area/tunnel free cross-sectional are

(a) (b) Fig 1. The grid meshes at the tunnel entrance and train head


Distance to tunnel entrance Train 0 m

150 m

500 m

350 m

700 m

Compression wave

Tail

Head

Expansion wave

0 2 4 6 8 10 12

P/kP

a

-4 0

4

t (s)

-4

0

4

0

4

Fig 2. Compression wave transportation process in tunnel (without shaft)

0 2 4 6 8 10-6

-4

-2

0

2

12

P (k

Pa)

t (s) Fig 3. Pressure fluctuation at train body (without shaft)

Fig 4. Compression wave transportation process in tunnel (shaft open ratio=14%)

-4

-2

0

2

4

P (k

Pa)

-4

-2

0

2

P (k

Pa)

0 2 4 6 8 10 12-4

-2

0

2

t (s)

P (k

Pa)

Distance to tunnel entrance -4 0 m

150 m

350 m

700 m

Compression wave

tail

head

Expansion wave

Shaft

Train

200 m


-2 0 2 4 6 8 10-4

-3

-2

-1

0

12

1

Shaft open ratio=14% Shaft open ratio=25%

P (k

Pa)

t (s) Fig 5. Effect of shaft open ration on pressure fluctuation at train body (with shaft)

Tab 2. Calculation Result Comparison

Maximum pressure at measurement point (kPa) Maximum pressure on train body (kPa)

Maximum pressure transient

P150 P350 P500

Working condition

positive negative positive negative positive negativepositive negative kPa /s kPa /3s

Without shaft 3.3 -2.9 3.2 -2.9 3.2 -2.7 0.3 -4.6 3.8 5.0

With shaft 3.2 -2.1 2.1 -3.2 2.2 -3.1 0.3 -3.4 3.0 3.1

The difference of aerodynamic effects with and without airshaft is shown in tab 2.

At the train body, the first peak pressure occurred the moment the train entered the tunnel. And the first negative peak pressure appeared when the train encountered with expansion wave converted from compression wave at tunnel exit. When compression wave passed through airshaft, energy was released, the peak value of the compression wave fell. From P150 to P350, the peak value of the compression changed from 3.2 kPa to 2.1 kPa. The reduction rate nearly amounted to 40%.

Figs 3 and 5 show that the compression wave at the tunnel exit induced the negative fluctuation at train body. By optimizing airshaft position to let the expansion wave pass through airshaft before encountering carriage, the peak value of expansion wave would be reduced. From this point of view, we proposed an expression to determine the optimal airshaft position:

shaft train tunnel tunnel shaft ,l l l l l

V c+ + −

>

Where, lshaft is shaft position to tunnel entrance, ltunnel is tunnel length, V is the speed of railway, c is the speed of sound, c=340 m/s. In this case, the optimal airshaft position is: 700 m＞lshaft＞235 m.

When compression wave passed through airshaft expansion wave formed， and then


encountered train body, causing negative pressure in cabin. Comparing fig 2 with fig 4, it is found that with airshaft, the moment first negative peak pressure appeared was ahead of the one without shaft, and the absolute value of peak negative pressure was larger than the pressure at 2.3 s in the case of no shaft. This is because the expansion wave transmitted to tunnel entrance and encountered with train body. When the train passed through airshaft, secondary compression wave was also generated. The expansion wave from the tunnel exit was reduced when meeting the secondary compression wave. In this way, the negative pressure on train body was lessened. In this case, expansion wave met train before it passed through airshaft. The peak negative pressure resulted from the expansion wave at the tunnel exit was almost equal the peak negative pressure resulted from the expansion wave passing through the airshaft. This validates that the choice of open area of the airshaft.

With the airshaft, the strength of pressure fluctuation was reduced, and the aerodynamic effects in cabin were moderated. In the calculation example, after setting airshaft, the peak pressure in three seconds, was reduced from 5 kPa to 3.1 kPa. The reduction rate is more that 40%.

IV. ENDING REMARKS

(1) The first negative peak pressure at carriage resulted from the train encountering with expansion wave, which was converted from compression wave at tunnel exit.

(2) By optimizing the shaft setting, the efficiency of ameliorating aerodynamic effects can be promoted.

(3) When a train passes through airshaft, secondary wave will be generated. The expansion wave at the tunnel exit can be reduced when meeting the secondary compression wave.

(4) With an airshaft, the strength of pressure fluctuation is reduced, and the aerodynamic effects in cabin are moderated.

ACKNOWLEDGEMENT

This paper is supported by the Fundamental Research Funds for the Central Universities (SWJTU09CX009).

References

[1]. R.S. Raghunathan, H.D. Kim, T. Setoguchi, Aerodynamics of high-speed railway train, Progress in Aerospace Sciences, 2002, 38(6-7): 469-514.

[2]. S. Mashimo, K. Iwamoto, T. Aoki, et al., Characteristics of compression wave generated by a high-


speed train entering tunnel, Engineering Sciences Reports, 1997, 18(4): 297-302.

[3]. S. Ozawa, Y. Moritoh, T. Maeda, et al., Investigation of pressure wave radiated from a tunnel exit, Tokyo: Railway Technical Research Institute, 1976 (in Japanese).

[4]. S. Ozawa, T. Uchida, T. Maeda, Reduction of tunnel exit boom by hood at tunnel entrance, Tokyo: Railway Technical Research Institute, 1977 (in Japanese).

[5]. S. Ozawa, T. Uchida, T. Maeda, Reduction of micro-pressure wave radiated from tunnel exit by hood at tunnel entrance, Quarterly Reports of the RTRI, 1978(19): 77-83 (in Japanese).

[6]. S. Ozawa, Studies of micro-pressure wave radiated from a tunnel exit, Tokyo: Railway Technical Research Institute, 1979 (in Japanese).

[7]. J.A. Fox, A.E. Vardy, The generation and alleviation of air pressure transients caused by the high speed passages of vehicles through tunnels, In: Proc. 1st ISAVVT, BHRA, 1973.

[8]. M. Tastsuo, Reduction of micro-pressure wave radiated from tunnel exit by branched in tunnel, Tokyo: Railway Technical Research Institute, 1977 (in Japanese).

[9]. A. Baron, M. Mossi, S. Sibilla, The alleviation of the aerodynamic drag and wave effects of high-speed trains in very long tunnels, Journal of Wind Engineering and Industrial Aerodynamics, 2001, 89(5): 365-401.

[10]. A.E. Vardy, The use of air shafts for the alleviation of pressure transients in railway tunnels, In: Proc. 2nd ISAVVT, 1976.

[11]. Y.X. Wang, B. Gao, K. Su, et al., Experiment research on reducing aerodynamics effect comprehensive measurement, Journal of Experiments in Fluid Mechanics, 2009, 23(1): 31-34.

[12]. Y.X. Wang, B. Gao, C.Q. Zheng, et al., Micro-compression wave model experiment on the high-speed train entering tunnel, Journal of Experiments in Fluid Mechanics, 2006, 20(1): 5-8.

[13]. R.S. Raghunathana, H.D. Kimb, T. Setoguchic, Aerodynamics of high-speed railway train, Progress in Aerospace Sciences, 2002, 38(6-7): 469-514.

[14]. M. Schultz, H. Sockel, Pressure transients in railway tunnels, W. Schneider, H. Troger, F. Ziegler, ed., Trends in applications of mathematics to mechanics, Harlow, 1989.

[15]. N. Komatsu, F. Yamada, The reduction of the train draft pressure in passing by each other, In: Proc. World Congrence Railway Research, Tokyo: Railway Technical Research Institute, 1999.

[15]. R.G. Gawthorpe, Pressure comfort criteria for rail tunnels operations, A. Haerter, ed., Aerodynamics and ventilation of vehicle tunnels, New York: Elsevier, 1991.

[16]. K. Sato, M. Ikada, M. Nakagawa, Effects of pressure changes on pain sensation of human ears, RTRI JNR,1989, 3(3) (in Japanese).

[17]. Y. Zenda, Study on the ventilating system of Shinkansen vehicle by simulating the internal pressure, RTRI JNR,1988, 2(12) (in Japanese).

[18]. Kobayashi, Suzuki Y, Akutsu K. Alleviating ear pains by controlling air pressure in ventilating system of Shinkansen car. RTRI JNR 1990, 4(7) (in Japanese). [19]. Fluent user’s guide, Fluent Inc. 2004-09♦


SERVICE LIFE ESTIMATION OF REINFORCED CONCRETE STRUCTURES WITH CONSIDERING THE DAMAGE

OF CONCRETE COVER

TRAN THE TRUYEN Department of Civil Engineering University of Transport and Communications

Summary: This paper presents an estimation of the long-term durability of reinforced concrete structures based on the rebar corrosion criteria. Service life is defined as the period of time required for the chloride concentration on rebar surface to reach the critical chloride level (Ccr). The corrosion rate of rebar is evaluated with considering the damage of concrete cover. The results show that when concrete cover is damaged, the service life of concrete structures is significantly reduced in comparison with the case, in which, concrete cover remains intact. In various exposure conditions of marine environment including spray zone, tidal zone, and atmospheric zone; the service life of concrete structures is different due to different corrosion rates of rebar.

Key words: Concrete, corrosion, rebar, durability, damage, time, chloride, diffusion, service life, reinforced concrete structures (RCS).

I. INTRODUCTION

Long-term durability of reinforced concrete structures (RCS) is affected by many physicochemical process. These processes include the chloride attacks, carbonation phenomenon, mechanical effects, etc. One of the most typical among them is the corrosion of rebar due to the penetration of chloride into concrete. The latter becomes serious when concrete cover is damaged in construction and in service due to mechanical effects. The damage of concrete first makes increase the seepage capacity of concrete cover, the chloride ion is then easily diffused into concrete causing damage of passive protection layer around rebar [8]. Once, the rebar is no longer protected, they are exposed to the attack of oxygen and water and finally becomes rusted.

The service life of RCS in this study is taken as the time from which the corrosion of rebar occurs due to the diffusion of chloride into concrete; or more accurately, this is the time from which, the chloride concentration C on the rebar surface reaches the critical value Ccr. The deterministic model used for the calculation of the service life is based on Fick’s second law [7]; the service life is then represented as a function of the surface chloride concentration Cs, chloride diffusion coefficient Kc, critical chloride level Ccr, and concrete cover thickness h. The service life is also effected by environments conditions; in this study, the marine environment conditions including spray zone, tidal zone, and atmospheric zone are considered. The


environment parameters in these zones included surface chloride concentration (Cs), chloride diffusion coefficient (Kc) is introduced based the studies of Mangat & Molloy (1994) [6] and A. Costa & J. Appeleton (1998) [3].

There is only the diffused damage state of concrete cover is considered in this study. Because, when concrete damage becomes localized or fully damage (fracture), the corrosion of rebar occurs immediately. Evaluation of RCS durability based on rebar corrosions criteria is no longer significant.

Effect of damage time of concrete on the RCS durability is considered in two cases: (i) Concrete is damaged just during the construction phases; (ii) Concrete is damaged after a period of putting into service.

II. PERMEABILITY AND CHLORIDE DIFFUSION OF DAMAGED CONCRETE

The relationship between gas permeability and chloride diffusion of damaged concrete is introduced based on the experimental results of Choinska & al (2008) [2]. After this research, the evolution of chloride diffusion with permeability of damaged concrete is represented in fig1.

Fig 1. Relationship between chloride diffusion and gas permeability of concrete

(Choinska & al (2008)) In fig 1, Kv(d) = K is the permeability of damaged concrete; Kvo = Ko is the initial value of

permeability (permeability of intact concrete): Ko ≈ 10-17 m2; Demig = Kc is the chloride diffusion coefficient into damaged concrete; and Demigo = Kco is the chloride diffusion coefficient into intact concrete. The relationship between chloride diffusion and permeability of ordinary damaged concrete (BO on fig.1) could be represented as a linear regression line with function as flow: Kc/Kco = 0.22(K/Ko) + 1 (1)

By replacing the permeability law of damaged concrete [10] in (1), we have: Kc/Kco = 0.22[α.exp(βD)] + 1 (2)

With the considered concrete M30 [10], (2) can be rewritten as follow: Kc/Kco = 0.22 exp(15.529D) + 1 (3)


1

3

5

7

9

11

0 0.05 0.1 0.15 0.2 0.25

D

K c /

K co

Fig 2. Increase of chloride diffusion with diffuse damage of concrete

The variation of chloride diffusion with diffuse damaged of concrete is shown on fig 2. We found that when the damage D < 0.1, the increase of chloride diffusion coefficient is weak. After this critical value, the chloride diffusion coefficient increases rapidly at reaches a values greater 10 times than the initial value.

III. SERVICE LIFE PREDICTION: NON-DAMAGED CONCRETE COVER

Concrete could be considered to be intact when the behavior of concrete are in elastic phase, i.e. no damage appears in concrete due to mechanical or temperature effects.

The model for evaluating the service life of RCS is developed from the equation calculating the concentration of chloride on rebar surface based on the second Fick law (RILEM 14 (2005) - Sara A.& E. Vesikari) [7]:

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛⎟⎟⎠

⎞⎜⎜⎝

⎛−=

tK2xerf1CC

csx

(4)

Where Cx is the chloride concentration at the depth x of concrete cover; Cs is chloride concentration on the surface of concrete structures; Kc is the chloride diffusion coefficient into concrete; t is the considered time; erf is the error function.

The corrosion process of rebar starts when Cx = Ccr; at this moment, x = h (concrete cover thickness), (4) can be rewritten as follow:

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛⎟⎟⎠

⎞⎜⎜⎝

⎛−=

tK2herf1CC

cscr (5)

The formula (5) could be simplified by using the Parabola function [7]:

2

cscr t3K2

h1CC ⎟⎟⎠

⎞⎜⎜⎝

⎛−= (6)

The chloride diffusion coefficient Kc and the chloride concentration as concrete surface Cs are time-depending factors.


To consider the variation of Kc, we can used the proposed model of Mangat & Molloy (1994) [6]: Kcot = Kco1.t-m (7)

Where, Koct is the chloride diffusion coefficient after t year; Koc1 is the chloride diffusion coefficient after first year; m is an empirical coefficient.

According to Mangat & Molloy (1994), values of Koc1 and m in different conditions of marine environments are different:

+ The spray zone: Kco1 = 3.12 x 10-12 m2/s; m = 0.51; + The tidal zone: Kco1 = 5.32 x 10-12 m2/s; m = 0.60; + The atmospheric zone : Kco1 = 1.21 x 10-12 m2/s; m = 0.42. Variation of surface chloride concentration Cs could be taken as the proposition of A. Costa

& J. Appeleton (1998) [3]: Cst = Cso. tn (8) Where, Cso is the surface chloride concentration after first year, n is an empirical

coefficient. Depending on the environment conditions, the values of Cso (% of concrete weight) and n

for ordinary concrete are different (A. Costa & J. Appeleton (1998) [3]): + The spray zone: Cso = 0.24; n = 0.47. + The tidal zone: Cso = 0.38; n = 0.37. + The atmospheric zone: Cso = 0.12; n = 0.54. If we consider the variation with time of chloride diffusion coefficient and surface chloride

concentration, the formula (6) can be rewritten as follow:

2

m-1co

nsocr

t3K2

h1tCC ⎟⎟

⎠

⎞

⎜⎜

⎝

⎛−= (9)

The minimum value of concrete cover thickness h to protect rebar from chloride attack is finally calculated by the following formula:

⎟⎟⎠

⎞⎜⎜⎝

⎛−= −

nso

crm1co .tC

C1t3K2h (10)

Fig 3 show the relationships between concrete cover thickness and the service life of RCS (Ccr = 0.06 % concrete weight [7]). We found that when concrete cover thickness increases, the service life is prolonged too. However, the evolution of the latter becomes slowly after a threshold of h. The environment conditions has significant effects on the service life. Indeed, at the atmospheric zone, the structure durability is the better; on the contrary, the tidal zone is the most dangerous zone for the deterioration of structures. Taking the case, in which, the concrete cover thickness equal to 6 cm: the service life of RCS is respectively about 80 years, 26 years, and 18 years at the atmospheric zone, at the spray zone, and at the tidal zone.


0

2

4

6

8

10

0 20 40 60 80 100 1 t (year)

h (c

m)

20

Tida l zone

S p ray zone

A tm osphe ric zone

Fig 3. Relation between concrete cover thickness and service life of RCS

IV. SERVICE LIFE PREDICTION: DAMAGED CONCRETE COVER

4.1. Concrete is damaged during construction phase

In this case, the chloride diffusion coefficient Kc is taken as the initial value Kco ≈ 5E-12 m2/s. The surface chloride concentration Cs is also the initial value Cso ≈ 0.35. The service life of RCS defined as the rebar corrosion criteria is then calculated by the following formula:

2

so

crco

CC

1

h]1)529.15exp(22.0[12K

1t

⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜

⎝

⎛

−+

=D

(11)

0

2

4

6

8

0 1 2 3 4 5 6

t (y e a r)

h (c

m) D = 0 .2 5

D = 0 .2 0

D = 0 .1 5

D = 0 .1 0

D = 0 .0 5

Fig 4. Effect of damage on the service life: concrete is damaged during construction phase

Fig 4 shows the influence of concrete cover damage level on the relationship between the service life and the concrete cover thickness h. We found that according to a value of the latter, the durability of structure decreases in consequence of the increase of damage level. The service


life is no longer significant when damage reaches the maximal value D = 0.25 (Threshold of diffused damage).

In comparison with the case, concrete is intact, we realized that the service life of RCS is strongly reduced even while the damage of concrete cover is still weak. Thus, the damage of concrete cover occurs during construction phase is extremely dangerous for the long-term durability of reinforced concrete structures.

4.2. Concrete is damaged after a period of putting structures into service

In this case, the chloride diffusion coefficient Kc is taken as the value just before the beginning of the damage of concrete cover: Kc = Kcot; Kcot is calculated according to (7). The surface chloride concentration Cs is also the value just before the damage of concrete Cs = Cst;

Cst is calculated by the formula (8).

In this study, we carried out an example of calculation with supposition that concrete cover is damaged after 5 years of putting structure into service. The service life is then calculated by the following formula:

5

5CC

1

h]1)529.15exp(22.0.[5.12K

1t

2

so

crco

+

⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜

⎝

⎛

−+

= −

n

m D (12)

Three areas including spray zone, tidal zone, and atmospheric zone are also considered as above.

Fig 5, 6, 7 show the relationship between the service life (calculated according to (12)) and the concrete cover thickness in different considered zones.

0

5

10

15

20

25

30

35

0 5 10 15 20 25 30 35 40 45 50 55

t (year)

h (c

m)

D=0.25

D=0.20

D=0.15

D=0.10

D=0.05

Fig 5. Relation service life – concrete cover thickness: Concrete is damaged

after 5 years in service (spray zone)


0

10

20

30

40

50

60

0 5 10 15 20 25 30 35 40 45 50 55

t (year)

h (c

m)

D=0.25

D=0.20

D=0.15

D=0.10

D=0.05


after 5 years in service (tidal zone)

0

5

10

15

20

25

0 5 10 15 20 25 30 35 40 45 50 55

t (year)

h (c

m)

D=0.25

D=0.20

D=0.15

D=0.10

D=0.05


after 5 years in service (atmospheric zone)

Results in figures 5, 6, 7 show that the damage state of concrete cover has significant impacts to the service life of RCS. Indeed, with the same concrete cover thickness, when damage increases, the long-term durability of RCS reduces rapidly. In the same meaning, to be able to have an expected durability, the concrete cover thickness should be increased when damage level increases. In comparison to the case that concrete is not damaged yet, the service life of RCS greatly decreases from the time that the damage process of concrete occurs.

The service life of RCS based in the tidal zone is always lowest in comparison with the value in atmospheric zone and spray zone.

III. CONCLUSIONS

The service life of reinforced concrete structures has been evaluated based on rebar corrosion criteria due to the chloride diffusion into concrete. The service life according to this criteria depends on the chloride diffusion coefficient Kc, the chloride concentration on concrete


surface Cs, and the concrete cover thickness h. When we suppose that the two first factors are unchanging, the service life is very low and doesn’t match with the real observation. By considering the variation of Kc and Cs with time, the calculated values of service life has reached the experimental values.

The marine environment conditions included atmospheric zone, spray zone, and tidal zone has significant effects on the service life of reinforced concrete structures. Indeed, the calculated results show that the latter is lowest at the tidal zone, and highest at the atmospheric zone. These results fully matche the mesured results of reinforced concrete structure durability in marine conditions.

The damage state of concrete cover strongly affects on the service life. Once, the damage level increases, the service life obviously decreases. Corresponding to the maximal value of diffused damage, the service life is no longer significant. The damage time has also strong effect on the service life. If the concrete cover is damaged during construction phase, the service life is very low, even no longer significant.

References [1]. Angst.U & al, COIN Project Report, Critical Chloride content in reinforced concrete, Coin workshop, Trondheim, Norway, June 2008. [2]. Choinska.M, Bonnet.S, Khelidj.A, Couplages endommagement – perméabilité – transferts d’ions chlorures, GeM, Université de Nantes, 2008. [3]. COSTA.A, APPLETON. J: Chloride penetration into concrete in marine environment - Part II: Prediction of long term chloride penetration, Materials and structures, 1998. [4]. Khatri.R.P, Sirivivatnanon.V, Characteristic service life for concrete exposed to marine environments, Cement and Concrete Research, 34, p745-752, 2004. [5]. Makeset.G, Vennesland. O, “Critical chloride content in reinforced concrete”, COIN Workshop, June 2008, Trondheim, Norway. [6]. Mangat, P.S. & Molloy, B.T. 1994, ‘Influence of PFA, slag and micro-silica on chloride induced corrosion of reinforcement in concrete’, Cement and Concrete Research, 21, 819–834 [7]. SAJIA.A, VESIKARI.E, “Durability design of concrete structures”, RELEM Report 14, 1996. [8]. SCHIESSL.P, Corrosion of Steel in concrete, RELEM-Report of the Technical Committee 60-CSC, 1998. [9]. Stanish. K.D., Hooton .R.D, Thomas.M.D.A, Testing the Chloride Penetration Resistance of Concrete: A Literature Review, Department of Civil Engineering, University of Toronto, FHWA Contract DTFH61-97-R-00022. [10]. TRAN T .TRUYEN, CHARLIER.R, “A method for evaluating the gas permeability of damaged concrete”, Science Journal of Transportation, No.03, Moscow-Chengdu-Hanoi, 11/2011. [11]. TRAN T. TRUYEN, TRAN V. TIEN, “Influence of marine environment conditions on the service life of reinforced concrete structures”, (In Vietnamese) Journal of Transport and Communications Research, No 32, Page 78 – 84, 6/2011. [12]. Zheng.j.j, Zhou. x.z, “Prediction of the chloride diffusion coefficient of concrete, Materials and structures, 2007♦


THE BLOCK OF MODELING AND FORECASTING OF GAS NETWORKS SYSTEMS OPERATION MODES

FOR DISPATCHER DECISION SUPPORT SYSTEM

BERNER, L.I. KOVALEV A.A. NIKOLAEV A.B. State Technical University – MADI, Moscow, Russia

Summary: The paper describes new methods of supervisory control automation aimed to help managers to solve the problem of maintaining a balanced mode transmission system. Forecasting the demand for gas and simulation modes of transmission system is complemented by a system of automated support for decision making.

This work was supported by the Government of the Russian Federation (Russian Ministry of Education) as part of the project under the Contract 13.G25.31.0064 on October 22, 2010.

Keywords: A system of automated support for dispatcher decision making, the task of balancing gas networks.

Maintaining a balanced mode of operation of gas transmission networks (GTS) is one of the most important tasks of supervisory control. The gas transportation system must have sufficient capacity to meet the demand of consumers for natural gas, even in cases of sudden fluctuations in demand caused by calendar, weather, economic or other reasons. The traditional solution of this problem is the usage of underground gas storage (UGS) and the gas stock in the tube as regulators. Gas suppliers can also introduce some regulators, but very often their ability to vary quickly the production volumes is limited by technology.

The task of balancing the severity of the regime is extremely important in cases of operation of long gas pipeline with a limited supply of gas in the tube. So far such pipelines were considered to be typical for foreign gas networks. However, in recent years in connection with the development of gasification of the Russian Federation and the construction of new modern pipeline (North - European, Sakhalin - Khabarovsk - Vladivostok pipelines, etc.), the problem is becoming more and more important for Russian gas industry. The solution of the problem is becoming even more urgent due to the development of market mechanisms of supply of gas, gas purchase on the "spot" markets, as well as due to climate variations, etc.

The paper describes some new automated dispatch control methods to help managers to solve the increasingly complex problem of balancing the GTS.


These methods have been worked out jointly by LLC "VIP" (the Russian subsidiary of the Group PSI AG, Germany) and JSC "AtlanticTransgasSystem" for their usage at the facilities of "Gazprom".

THE TASK OF BALANCING THE CTA AND DECISION-MAKING

Fig 1 shows the pipeline, which is the subject of balancing.

Fig 1. The structure of the system balancing

The gas enters the system from vendors (fisheries) and other companies to transport gas, and can also be selected from the storage facilities where available. The gas is supplied from the system to consumers, other companies (gas transit), is pumped into underground storage facilities, as well as is consumed for their own needs (the work of gas pumping units and other technological problems). The pipeline itself functions also as storage of gas. As a result, the supply of gas in the pipeline may both increase and decrease. The task of balancing the presented system in principle is obvious - the gas must flow into the system in an amount sufficient to meet consumers' needs, personal needs, as well as for further transit. For various reasons, the demand for natural gas may fluctuate. First of all, the demand depends on weather conditions and, often, on the calendar date - the day of the week, holidays/workdays, etc.. A proportion of demand is associated with spontaneous or predictable seasonal fluctuations in economic activity. For example, in agricultural regions the plants consume gas is only a few months of the year for the processing of the crop. Unfortunately, most part of the demand fluctuation is much less predictable. It should be noted that taking into consideration the length of the pipelines, the pipeline suppliesnatural gas to consumers in areas with different weather conditions, which complicates the forecast of demand fluctuations.

Assessing the situation and determining either the excess or deficiency of gas in the system, the dispatcher must decide on balancing "supply and demand". Possible actions may include:

• Selection or gas injection from / to storage facilities (if any);

• Increase or decrease the supply of gas in the tube (ie, injection of gas into the pipe or the selection of the gas pipe - in fact, an analogue of UGS smaller scale);

• Request for additional volumes of gas from suppliers or the opposite - a request for reduction of gas supplies;


• In an extreme case - failure / restriction of additional or "spot" applications for the supply of gas and very emergency - the restriction of some consumers.

Thus, in order to balance the system it is necessary to settle two main questions: (1) what is the current situation and trends of its changes - is there enough gas in the near future, (2) what measures from those listed above to take in order to align the imbalance.

Taking into consideration the inertia of the gas transmission system and of storage facilities, a prerequisite is to predict imbalances for the period of time, which at least allows the dispatcher to take effective measures to eliminate potential problems.

DISPATCHING ORGANIZATION ALTERNATIVE

The following solution is proposed for organizing and controlling the gas transportation system to balance the supply of gas, as well as for solving problems of safe operation of the pipeline and the optimization of the equipment. In addition to the basic system of supervisory control (SCADA PSIControl), "on-line" model of the pipeline (PSIGanesi), «off-line" model, a system of forecasting gas consumption, and the newly developed PSIPrognosis software system analysis and decision support (SPPDR) are realised.

Fig 2. The structure of the system (supply)

A generalized algorithm SPPDR shown in fig 3. Modeling of the "future", based on the targets is a tool to identify imbalances in the system.

Analysis of the results of modeling is performed by the dispatcher "manually", as well as with the help of automated mechanisms for expert systems - SPPDR module. Heuristic rules SPPDR allow the dispatcher to make recommendations - to change the mode of the compressor station (CS), request a change in the regime of fishing, to change the supply of gas in the tube, away from the gas storage facilities, restrict consumers. Recommendations SPPDR can be considered to be preliminary. Optimal solutions for the pipeline mode of operation should be obtained interactively using the gas-hydrodynamic model of "Astra." Model "Astra" calculates various options for MG modes, which on the advice of an expert system SPPDR or proceeding from its own considerations, the manager offers. Option is selected, or the best of the fuel and energy


costs, or preferred by some subjective reasons.

Fig 3. The control algorithm (proposal)

The final choice of options for changing the operating mode of the COP and the MG as a whole takes the dispatcher. The selected option mode takes the form of supervisory tasks and settings for transfer to lower-level automation system.

The proposed approach implements not only remote but also the intelligent management of the pipeline mode of operation and allows quicker and with less expenditure of energy to balance the important pipeline system. Domestic consumers and, in the future, foreignconsumers are provided with a guaranteed supply of gas.

In addition to the problem of balancing work of MG and ensuring guaranteed delivery of normal modes, the system can be used in emergency situations, helping to identify alternative ways of organizing supply and install a real need of limiting gas consumption.

CONCLUSIONS

This paper proposes a modern and perspective, according to the authors, method of organization of supervisory control and solving an important task of balancing modes of gas transmission system. The authors hope to put into practice the developed (at the level of a model) idea and apply it to provide efficient and reliable operation of gas transmission networks.


INDUSTRIAL INTEGRATION AND ECONOMICAL CRISIS

GALINA BUBNOVA Moscow State University of Railway Transport Russia

Summary: Taking into consideration the dominance of the feed component in the tradable mass produced by the majority of Russian enterprises and destabilizing tendencies on the world’s markets, the task of organization of tolling business with the participation of major shipping companies becomes actual. In the given context the differences between available technology of domestic and foreign companies, their productive capacity, volume and quality of reprocessing of the primary product, necessary volumes of resources of the primary products to ensure the steady production in Russia are taken into consideration.

The Business participation of the shipping holding company PLC ‘Russian Railway’ in organization and support of the tolling business does not only broaden the bounds of its business, but it is also a long-range tendency of decreasing the painful consequences of the international economic crisis.

The analysis of the Russian industrial enterprises with tolling schemes allows us to reveal problems that limit the effectiveness of its application – these are the legitimacy of application of such operations, cohesion and uncertainty of bounds of tolling and customer-owned schemes.

The absence of the definition of tolling on the legislative level leads the experts in this field to multiple-valued conclusions regarding its application and the development of these operations.

According to the definition of the International Trade Centre (UNCTAD/WTO) a tolling contract is a variant of the take-or-pay contract, which means that the condition of contractual obligations is unconditional purchase of a product or a service offered to the purchaser with the condition of buying outright. In the case of refusal of the purchase the buyer must pay the seller an appointed sum of cash means. In the case of the tolling contract it is the payment for the usage of the infrastructure objects, and in the case of the take-or-pay contract it is also for the non-fulfillment of the contractual obligations imposed on one of the participants.

The Russian equivalent of the mechanism of the tolling truck represents an operation of reprocessing of the customer-owned primary product and materials. Mutual relations, based on such schemes, correspond to the relations of rendering of services provided by specialized enterprises such as fulfillment of functions of maintenance supply of one enterprise and marketing in the bounds of the existing production process. The primary product and components act as customer-owned primary product, they are passed from the owner of production to the manufacturer for the purpose of their completion or reprocessing into a


finished product and, by that, they represent the basis of the tolling truck.

Why such relations are substituted for simple operations of customer-owned primary product industrial enterprises in a number of cases? To a great degree it depends on the practice of using the conception of the reprocessing of the customer-owned primary product in tolling schemes during the times of formation of aluminium production. There is still no definition of the concept of tolling in the Russian law, but there are characters based on which the operations of industrial enterprises, that work on customer-owned primary product, are labeled as such.

In general, the roles of the participants of the tolling truck are acted by such subjects as: the owner of the primary product, the reprocessor of the primary product, the owner of the finished product (the result of the reprocessing), the consumer and the middlemen.

The indication for labeling the requestor of reprocessing as a middleman during the realization of the tolling schemes coincides with the indication of transition of the proprietary right for the results of the reprocessing of the primary product and materials to the client. In this context it means that the corresponding proprietary right has been passed to him/her according to the sale agreement of primary products and materials. This indicated that the tolling middleman acts as a specialized company that combines the functions of maintenance supply for the reprocessor of the primary product, on the one hand, and, on the other hand, the marketing function for the primary owner of the primary product and materials for the further reprocessing in the higher redistribution of the products (fig 1).

Fig 1. The diagram representing production on the customer-owned material

The supplier of primary

product

The reprocessor of the primary

product

The tolling middleman

The

consumer

1

2

3

4

with participation of the tolling middleman

During the realization of this type of the tolling scheme the primary owner of the primary product becomes the supplier of the primary product for the tolling middleman, the proprietary right for the result of the reprocessing belongs to the middleman until the primary product is given to the reprocessor. In the end the finished product is distributed to the consumers by means of the tolling middleman. Organizationally, the participants of this scheme are not connected, and in some cases the tollinger acts as a financial middleman.

In general the scheme consists of the following steps:

1. The supplier passes the proprietary rights (sells) to the primary products or materials that


are at his/her disposal, because he/she does not possess the necessary and/or free facilities for its reprocessing, or the passing product is the final step in his/her production process;

2. The tolling middleman, acting as a specialized company, that does not possess any specific equipment for the reprocessing of the acquired primary product, sends it for reprocessing to an enterprise that possesses the right technological equipment or available equipment for this purpose;

3. In accordance with the results of the reprocessing of the primary products and materials the ready product of higher redistribution is passed to the tolling middleman;

4. The tolling middleman sells the finished product to the customers.

It should be noted that in some cases tolling is viewed as a variant of counter trade that can be placed in-between barter operations, trade compensation and operations of industrial compensation. In this case the scheme is completely different, the role of the tolling middleman is passed to the owner of the primary product and materials, and the owner passes them to the reprocessor of the primary products (1) for the purpose of acquisition of products of higher redistribution (2) and, later, the owner sells it to the customers (3) (fig 2).

Fig 2. Diagram representing production on the customer-owned material

The reprocessor of the

primary product

Primary product

owner

The consumer 1

2

3

Δ1 or Δ2

The fundamental difference of the tolling scheme from such forms of organization of the distributed production consists in the fact that tolling can be characterized by saving the proprietary right for the products that are received in 100% volume from the received primary products or materials and it does not make a provision for non-monetary compensation of the reprocessing services (Δ1 or Δ2). However it is stipulated that a part of the primary product or material or a part of the finished product is given to the reprocessing enterprise as a payment for the reprocessing services.

We find the explanation of the specificity of the mutual settlements in Russia during the period of cash shortages characterizing the economy in the beginning of the 1990s as an argument for calling tolling a barter type operation. Classification of these types of schemes as an operation with the customer-owned material left traces and has common features with the tolling schemes in question.

Tolling schemes (fig 1) started their development by putting the requestor of the reprocessing in the middleman category in realization of such schemes, acting as a collaborating link in the organization of divided production and was also the owner of the primary product (materials). Being such a ‘middleman’ the reprocessing client performed in this scheme such functions that the primary owner of the primary production did not have. Such functions were:


1. Maintenance supply for the reprocessor of the primary product;

2. Marketing functions for the primary owner of the primary product.

If we proceed from the practice that existed when the owner of the primary product was also the requestor of the reprocessing (both with monetary and barter (non-monetary) types of settlement), then this variant is not being used nowadays, but it cannot be excluded during the times of the economic crisis.

The maintenance supply problem is resolved only in part also because the owner of the primary product cannot solve it on his own, having a limited quantity of resources for the reprocessing which cannot provide full utilization of the available productive capacity of the reprocessing enterprise. Besides, the quality of the supplied primary products and materials must satisfy the technical requirements of the processing enterprise.

The insufficiency of such scheme reveals itself in the fact that the client solves his/her own problems, particularly the inability to reprocess given primary products and/or materials. The economical point of this scheme is justified - the finished higher redistribution product made of one’s own primary product has much higher value which has a good effect on the financial result of both the processing enterprise and the requestor of such reprocessing.

The limits of the usage of the scheme are obvious, it solves problems of only one participant in an industrial relation and does not have the same selection of functions that would allow, as a result of such cooperation, to create a larger surplus value which permits to solve the tolling schemes.

Therefore, unlike all the other forms of lean production (including production as a part of agreements concerning division of production and unreliable transactions) tolling does not substitute for barter schemes in the settlement between the parties. Also tolling does not substitute for partial settlement with the finished product for the services of reprocessing of the customer-owned materials (primary products) to the reprocessor of the primary products, which is reflected in the current Russian practice.

Since 2006 questions of the tolling production were included in the list of pressing problems by a group of experts in national income accounting under the UN Economic and Social Council (ECOSOC). Following the results of the first meeting in April of 2008, which was held with the assistance of the Economic Commission for Europe, Statistics Unit of the EU, Organization for Economic Cooperation and Development, Working Group on the Impact of Globalization on National Accounts (WGGNA) special attention was paid to the tolling production and reprocessing of goods.

The results of the influence of tolling schemes, that are part of a special-purpose chain of the delivery of goods, will be reflected in June 2010 during the conference of European statisticians in the outcome document of the WGGNA.

According to the WGGNA the subject matter of tolling is as follows:

- The proprietary rights on goods are kept by the supplier and are also extended to the


products of reprocessing;

- The reprocessor is paid for the work;

- The enterprise sends the goods to the other country with the purpose of reprocessing keeping the proprietary rights on these goods which will later extend to the products of reprocessing that will later be sold in third countries.

This definition corresponds to the international chain of product deliveries which also specializes on the tolling production. Consumers buy the proprietary right from the enterprise which sent the goods to reprocessing. This procedure is characterized by the difference in the quantity of the surplus value between the reprocessed goods and the goods accepted for reprocessing and which cannot be determined in advance by the supplier. The beneficial economic effect attracted the interest of different participants of foreign economic activity, which was the reason for the European Economic Commission’s attention to this process. The ultimate aim of organizing tolling production for international chains of goods delivery will be to supply the entire chain with steady production processes and at the same time observe the necessary balance of interests among all the international specialization participants and to promote cooperation during international trade.

Taking into consideration the UN definition of reprocessing of goods and tolling production, the necessary condition for tolling is direct participation of parties of such an agreement in the foreign commerce turnover during trade through the customs border of countries. This calls for observance of routine of the movement of goods through the customs borders of countries.

The existing legislation of Russia contains the rules according to which manufacturing or other activities of the subjects of foreign-economic activity (international trade) is classified by the experts as tolling. In general tolling relations are regulated at the levels of existing norms of Russian and international law, even though these rules and norms are not directly described as tolling relations.

The customs code makes provisions for the following customs regime which are applied to the goods for which reprocessing is provided:

- Reprocessing in the customs territory;

- Reprocessing for domestic consumption;

- Reprocessing outside of the customs territory.

Customs regimes that are described in the latest editions of the Customs Code are in content similar to procedures applied in accordance with the International Convention for Simplification and Harmonization of Customs Procedures. Since this international document was not ratified we will make an analysis of procedures of customs in reference to the criterion for labeling foreign trade dealing as these regimes with the purpose of specification of tolling schemes, the results of which are shown in the tab 1.


Tab 1

Criterion Reprocessing in the customs territory

Reprocessing for the domestic consumption

Reprocessing outside of the

customs territory

Customs territory

Russian Federation Russian Federation Other countries

The requirements on the term of reprocessing

Not more than 2 years Not more than 1 year Not more than 2 years

The conditions of customs regime

Reprocessing of goods with further export of products of reprocessing from the customs territory

Reprocessing with further release of reprocessing products for the free circulation

Reprocessing of goods with further import of reprocessing products

Economical cosequences

Full conditional exemption from the payment of customs duties and taxes

Full conditional exemption from the payment of customs duties till the release of the products of reprocessing for the free circulation, the payment of customs duties for the reprocessing products afterwards.

Full or fractional conditional exemption from the payment of customs duties and taxes

The customs regime of reprocessing for domestic consumption is not being applied at present, because permissions for reprocessing are not issued until the Government of the Russian Federation (according to p.2 section 188 of the Customs Code of the RF) has ratified the list of goods as regards to which the reprocessing for domestic consumption is allowed.

Analyzing reprocessing regimes in the customs territory and reprocessing regimes outside the customs territory we have discovered that the main difference is in the customs territory, in the country where the reprocessing of goods occurs. Accordingly, for the first regime such country is the Russian Federation and for the second regime it is another country. In 2008 the ratio of regimes was 1:13, fig 3.

93%

7%

Reprocessing in the customs territory

Reprocessing outside of the customs territory

Fig 3. The share in the cost estimation of goods of customs regimes of reprocessing

Reprocessing on the customs territory is a regime under which the imported goods are used in the customs territory of the Russian Federation during the time-frame (the time of the reprocessing) for the purpose of executing operations of reprocessing goods with full conditional exemption from the payment of customs duties and taxes with the condition of taking out the results of reprocessing from the customs territory of the Russian Federation during the given period (fig 4).


Fig 4. Diagram representing the reprocessing regime in the customs territory

The reprocessor of the primary product

Tolling

enterprise

The

consumer

1

2

3

Russian Federation

According to the content of the reprocessing regime in the customs territory, the goods that were imported for the purpose of reprocessing (1) are liable to obligatory export from the customs territory of the Russian Federation as reprocessing results (2) and in the end the can be sold to consumers (3). This regime will be reclassified as a regime of reprocessing for the domestic consumption with all the economic consequences, if the conditions of placing the goods under this customs regime (such as identification of exported goods as products of reprocessing) are violated.

In the category of goods can be placed any moveable property that is moved across the border, and for the purpose of reprocessing primary products and materials can be classified as such goods.

In accordance with the customs regimes during the organization of tolling schemes the following operations of reprocessing goods are possible:

1. Proper reprocessing or processing of products;

2. Manufacturing of new products, including assembling, installation and disassembling of products;

3. Epairs of goods, including their restoration, changing of their components, restoration of their essential properties.

In 2008 the largest part from the goods placed under the reprocessing customs regimes were received by equipment and mechanisms in group of reprocessing in the customs territory regime and transport in group of reprocessing outside of the customs territory (tab 2).

Tab 2

The name of the heading (the group of goods)

Reprocessing in the customs territory

Reprocessing outside of the customs territory

Total

1. Equipment, mechanisms 40,68% 10,23% 38,55% 2. Plastic products, rubber 16,54% 1,14% 15,47% 3. Base metals and products 13,26% 6,20% 12,77% 4. Transport 4,59% 75,57% 9,54% 5. Textile 7,90% 0,18% 7,37% 6. Chemical products 5,75% 0,37% 5,38% 7. Optical devices and instruments 4,95% 2,01% 4,75% 8. Mineral products 2,93% 4,05% 3,01% 9. Wooden, paper and cardboard mass 1,91% 0,01% 1,78% 10. Different industrial goods 1,48% 0,24% 1,39% Total 100,00% 100,00% 100,00%


In the joint group of goods without division into customs regimes the following are prevalent: equipment and mechanisms, plastic and rubber products and, of course, base metals and products made from them. The placed orders in these groups prove that the reprocessing enterprises are well equipped with productive assets and a relatively inexpensive labor force.

The materials, primary products and goods received for the reprocessing are characterized by a moderate level of redistribution. The resulting products of reprocessing will be components or semi-finished product for another enterprise. And the high level of ‘foreign’ good that are reprocessed speaks about the inapplicability of its usage in domestic production, and in case of the impossibility of completing reprocessing independently these products should be ordered in other countries where they are applicable, as in case of goods of the transport category (75,57%).

The content of this customs regime reflects the essence of the tolling scheme and its difference from other forms of lean production, which assumes non-monetary compensation of reprocessing services, and the differences are revealed in the following:

- The conditions of exporting products of reprocessing; - The relations of leftover goods that were placed under the ‘in the customs territory’

customs regime. The conditions of exporting products of reprocessing (in particular those that are contained

in p.2 section 178 of Customs Code of the RF) state that the description, quality and quantity of reprocessed products are finally defined after the reconcilement of the reprocessed products yield norm, which, in their turn, are reconciled with customs during the expert and/or laboratory appraisal based on the specific technological process of reprocessing.

In respect to the leftover remains of the goods, that can be used as a payment means for the services of reprocessing in barter type deals, these goods are characterized by the customs clearance charge and are not subject to declaration.

The remains of goods placed under the customs regime of reprocessing in the customs territory can be exported from the customs area of the RF without customs clearance charge or can be placed under the customs regime of reprocessing in the customs territory.

The dispositions of these norms of the law exclude every type of compensation deals or barter deals the basis for which can be the reprocessing of the customer-owned primary product (material), but that do not fit in the definition of the tolling scheme according to the UN version, and in particular remission of customs duty. This fact makes participation in tolling schemes one of the arguments for the attractiveness of such a form of business partnership.

Taking into consideration the legislation in force, tolling is excluded from the list of barter business, therefore tolling enterprise cannot be viewed as a commercial middleman in unreliable barter transactions.

Total exemption of customs clearance charges will be executed under the condition of exporting the products of reprocessing from the customs territory or their importing in accordance with the conditions of customs regimes.

Keeping in mind the results of the analysis of principles of tolling organizing tolling schemes and the contract relations connected to them, we have a fundamentally new scheme of cooperation of economic entities, in conformity with the recent requirements of the existing legislation of the RF in the field of international trade (fig 5).


Fig 5. The diagram of production on the customer-owned primary product (tolling scheme)

The consumer

The primary supplier

Tolling enterprise

Primary product owner

The reprocessor of the primary product

The owner of the

finished manufactures

1.2

1.1

On the diagram the home manufacturing is centered in the hands of the reprocessor of the raw material when supply and sales functions are distributed among other subjects of the tolling scheme. This tolling scheme unites the following participants:

1. Tolling enterprise: 1.1. The owner of the primary product; 1.2. The owner of the finished product; 2. The reprocessor of the primary product; 3. The original supplier; 4. The consumer. Acting as a requestor of reprocessing tolling enterprise acts as the owner in respect to the

supplied primary products and materials. During the reprocessing under conditions like tolling, the proprietary right on the products of reprocessing is kept by the owner of the primary product and materials.

On this diagram the division of the tolling enterprise is done under the condition that the customs regime of reprocessing for domestic consumption (the regime is allowed, but does not yet exist) may be joined into force, because at present there is no direct law that prohibits that. In this case the products of reprocessing are exempt from exportation, and the subsidiary production unit (that exists as a part of cooperating tolling enterprise) should act as the owner of the finished products. In the legal context it means that being a foreign company, that supplies the primary products and materials for the purposes of reprocessing, this company (under the regime of reprocessing for domestic consumption) should also act as the owner of the finished manufactures in the Russian Federation and should have all the taxpayer’s right and responsibilities connected to the transition of goods through the customs borders of countries.

In this case the subsidiary production unit of the tolling company together with the registered outside capital represents its interest on the territory of the Russian Federation. Therefore, the functions of the tolling enterprise are divided according to the territorial mark connected with the allowed reprocessing regimes with the condition of keeping proprietary


rights on the conversion product. The functions of the tolling enterprise are not reduced to the function of a middleman in the

tolling schemes. Middlemen are singled out in an independent category, and the transport enterprises act in this role. It is defined according to the following:

1. The realization of logistical aims: - Marketing function for the primary owner of raw products and materials; - Maintenance supply for the reprocessor of the primary product. 2. The aims of managing ‘just in time’ commodity flow (assuming the fixed term of

reprocessing goods) may be accomplished only by a specialized company, the function of which is given to a tolling enterprise for outsourcing. Under the conditions of cooperating for tolling production, the transfer to outsourcing is an economically justified decision within the limits of production processes of distributed production with the reconcilement of goods supply chains.

The tolling enterprise, accomplishing the aims of logistics on its own without the involvement of transport enterprises, is not capable of accomplishing them measurably, and in the opposite case the independent accomplishing of these aims and functions is more extravagant.

Summarizing the results of the conducted analysis, let’s point out the most important, to our mind, advantages for the participants of the tolling schemes.

For the suppliers of primary products and materials: tolling relations will allow them to broaden the range of launched products as a result of a magnified number of reprocessing enterprises capable of increasing not only the quantity of launches, but are also capable of improving the quality of the mass of goods.

The basis of tolling enterprises is balancing the interests of participants of such schemes and relations under the conditions of managing the flow of primary products, materials and the reprocessing products, so the tolling enterprises will be ensured of the distribution of such technologies among the regions and also penetrating new markets. It is obvious that the advantages and profits of such integration will be applied to all the participants of tolling relations.

To the reprocessor of the primary product tolling schemes will allow ensuring the full load of its productive capacities without the involvement of additional sources of financing for its own productive program and without a substantial loss of economical profits in the future. At the same time the enterprise that is provided with the utilization ratio indirectly solves the social problem in a town where this enterprise is strategic.

For the transport company (PLC ‘Russian Railway’) shared participation in the tolling enterprise business is one of the most promising areas of focus of the business development of PLC ‘Russian railway’. The tolling enterprises in question are those that are profiled for the nomenclature of primary products and materials that are used in transport production or that are in demand for production of other goods that are in their turn necessary to ensure the transport production.

The tolling enterprises as international integrators can serve as a peculiar kind of ‘a pillow’, the amortisseur of painful strokes of the economic crisis gathering pace in the different countries of the world.


INITIAL RESULT OF USING CINDER PARTICLES AT SOME STEEL FACTORIES IN BA RIA - VUNG TAU PROVINCE

AS A MINERAL ADDITIVE FOR CEMENT CONCRETE IN BUILDING MOTORAY SUFRACE

DR. LE VAN BACH University of Transpor and Communications ENG. TRAN HUU BANG Saigon Construction Quality Control Jont Stock Company

Summary: The report presents initial result of studying characteristics of cement concrete using slag as a mineral additive at some factories in Ba Ria - Vung Tau province to manufacture cement concrete for construction of motorway. This application is, in fact, piloted in some road sections.

I. EVALUATION OF SLAG PARTICLES AT SOME STEEL FACTORIES IN BA RIA - VUNG TAU PROVINCE

Steel slag is waste in metallurgy, which is used as a waste product in metal production process from iron ore or unpure metal refining process. Iron ore normally contains clay and sand, so people usually add appropriate content of limestone to the furnace during production. During burning, iron ore and limestone react with each other to form silicate calcium, alumina silicate and aluminate silicate and magnesia calcium. Steel slag is melted at 1400 - 1600oC. At this temperature, compounds are completely melted. The specific weight of melted compounds is smaller than that of cast iron, so it floats on top. People remove and call this product slag. Steel slag has various forms depending on burning process and cooling mode after melting. Quick cooling is accompanied with limited quantity of water, then the steam is withdrawn, leaving pore, honeycombed in slag structure similar to holystone. Light foamed slag is then crushed and used a mineral additive for concrete.

Ba Ria - Vung Tau where there is concentration of many steel factories is considered as metallurgy center in the South with output of 3.75 million bloom yard ton/year, forecasted weight of 412,000 - 562,000 steel slag ton/year. Steel slag arising from steelmaking which isn’t recycled generates dust, overflown rainwater through slag storage. This will be a burden on environmental and economic issues for steelmaking plants. At the same time, it will lose land and waste resources if being dumped (Source - Green Resources Joint Stock Company, 24/06/2011).


Figure 2. Steel slag storage at Toc Tien commune, Tan Thanh district, Ba Ria – Vung Tau province

Figure 1. Steel slag storage of steel factory in the South, Ba Ria - Vung Tau

II. SOME TESTING RESULTS IN CRITERIA OF CEMENT CONCRETE USING WASTE STEEL SLAG AS A MINERAL ADDITIVE

2.1. Preparation of materials for testing - Cement: Holcim PCB40 cement. - Sand: Dong Nai river sand (type of sand used to produce cement concrete). - Rock: exploited and manufactured in Hoa An stone quarry, Dong Nai province. - Water: water for conventional concrete. - Mineral additive: Steel slag particles have form of black grey clots, spongy volume

weight of 900kg/m3, specific weight of 2.5 g/cm3.

Figure 3. Steel slag sample before and after being crushed to produce a mineral additive

- Chemical components of steel slag: SiO2: 28 - 38 %; Al2O3: 8 - 24 %; Fe2O3: 0,64%; CaO: 30 - 50 %; MgO: 5 %; S: 1 - 2.5%.

- Mineral components of steel slag:

Ghilenit (2CaO. Al2O3.SiO2, CaO.SiO2, 2CaO.SiO2) mineral component. In addition, there are Monticenit (CaO.MgO.SiO2), Akemanit (2CaO.MgO.2SiO2), Merwinit (3CaO.MgO.2SiO2), Anorthit (CaO. Al2O3.2SiO2), Spinel (MgO.Al2O3), Fortenit (2MgO.SiO2) và các Aluminate canxi (CaO.Al2O3, 12CaO.7Al2O3).

Design of conventional cement concrete aggregate components and cement concrete uses steel slag mineral additive (table 1).

Tested cement concrete grades: 25MPa; 30MPa and 35MPa.


Table 1. Calculated results for conventional cement concrete grade components Material component for concrete (1m3) Concrete grade component X (kg) C (kg) Đ (kg) N (kg)

Design humidity W=0%. 25Mpa grade 353 754.9 1098 205 Adjustment of humidity of Wđ = 0.29 % and Wc = 7.35%. 25Mpa grade 353 810 1101 147

Designed humidity W=0%. 30Mpa grade 404 722 1085 205 Adjustment of humidity of Wđ = 0.29 % and Wc = 7.35%. 30Mpa grade 404 775 1088 149

Design humidity W=0%. 35Mpa grade 468 669 1065 211 Adjustment of humidity of Wđ = 0.29 % and Wc = 7.35%. 35Mpa grade 468 718 1068 160

Note: X - cement; C - sand; Đ - rock; N - water. Design of cement concrete uses steel slag mineral particles: based on design steps of

conventional concrete. But replacement of cement content is in proportion to percentage: 10%, 12% and 15% slag in mixed components of the aggregate. Results are shown in table 2 after adjusting humidity of Wđ = 0.29 % and Wc = 7.35%.

+ Age of tested cement concrete sample: - Sample for testing compressive resistance Rn applicable to 5 periods of days: 7, 14, 28, 60

and 90; - Sample for testing tensile resistance Rku applicable to 2 periods of days: 28 and 60. + Sample dimensions: - Compressed sample: 15x15x15 cm; - Tensile sample: 15x15x60 cm.

Table 2. Design of concrete grade components using steel slag as a mineral additive

Materials components for 1m3 cement concrete with mineral additive

Components of cement

concrete aggregate X (kg) C (kg) Đ (kg) N (kg) Steel slag

10% (kg) Steel slag 12% (kg)

Steel slag 15% (kg)

317.50 810 1101 147 35.50 310.64 810 1101 147 42.36 25Mpa 300.05 810 1101 147 52.95 363.60 775 1088 149 40.40 355.52 775 1088 149 48.48 30Mpa 343.40 775 1088 149 60.60 421.20 718 1068 160 46.80 411.84 718 1068 160 56.16 35Mpa 397.80 718 1068 160 70.20

+ Number of samples to be manufactured: - Types of sample for testing compression 15x15x15cm: Number of samples using steel slag: 3 slag grades x 5 types of age x 3 grades x 3

samples/group = 135 compression samples; Number of check samples (without steel slag): 5 types of age x 3 grades x 3 samples/group = 45 compression samples.

- Types of sample for testing tensile resistance 15x15x60 cm:


Number of samples using steel slag: 3 slag grades x 2 types of age x 3 grades x 3samples/group = 54 samples; Number of check samples (without steel slag): 2 types of age x 3 grades x 3 samples/group = 18 samples.

Figure 4. Cubic compression samples 15x15x15cm and tensile samples 15x15x60cm

Figure 5. Testing for compression and tensile testing for cement concrete containing 15% mineral additive

Some remarks: - Cement concrete strength containing steel slag mineral additive develops more slowly

than conventional cement concrete. - Next study results after 28 days show that in age of long-time when minerals in Portland

cement have nearly liquefided, components in steel slag continue to react with Ca(OH)2 to form C-S-H products, to increase condensation and strength for cement concrete containing steel slag mineral additive to approximate strength of concentional cement concrete. (For example, after 60 days, compression strength of conventional cement concrete compared to concrete containing 10%, 12% and 15% mineral additive in term of 30MPa sample is, in turn, 35.9, 35.0, 34.2 and 32.7MPa - table 3.

Table 3. Compression strength of cement concrete with and without steel slag Compression strength in different ages of days (MPa) Sign of sample R7 R14 R28 R60 R90

Conventional 25MPa 19.6 23.3 28.1 30.1 30.8 25Mpa; 10% steel slag 18.9 22.6 27.5 29.0 30.5

12% steel slag 17.8 21.7 26.5 28.5 29.5 15% steel slag 15.9 19.8 25.2 27.0 27.2

30MPa, conventional 23.8 27.4 33.5 35.9 36.1 30MPa; 10% steel slag 22.5 26.5 32.5 35.0 35.8


35MPa, conventional 26.5 31.7 38.5 40.6 40.7 35 MPa; 10% steel slag 25.6 30.5 37.2 39.4 40.4



Table 4. Tensile strength of cement concrete with and without steel slag when bending Tensile strength (daN/cm2) Sign of sample R28 R60

25MPa, conventional 33.51 34.31 25 MPa, 10% steel slag 32.58 33.64

12% steel slag 29.38 30.22 15% steel slag 28.31 29.02

30MPa, conventional 38.97 39.78 30MPa, 10% steel slag 36.28 37.73


35MPa, conventional 40.60 41.56 35MPa, 10% steel slag 38.91 40.40


Figure 6. Diagrams for comparing tensile strength

of cement concrete f 25, 30 and 35 MPa grades with and without steel slag in different ages of days

III. PILOTING FOR CONSTRUCTION OF A ROAD SECTION BY CEMENT CONCRETE WITH STEEL SLAG

3.1. Introduction to the project

- Project: Internal road.

- Location: My Xuan commune, Tan Thanh district, Ba Ria - Vung Tau province;

Technical specifications of the road:


- Length: 430m;

- Road surface width: 9.2m;

- Concrete slab thickness: 15cm;

- Concrete slab width: 4.58 x 3.0m;

- Expansion joint width: 15mm;

- Cement concrete grade: 25MPa.

- Percentage of steel slag mineral additive: 12%.

The employer: My Xuan brick and tile joint stock company;

Concrete supply unit: Development Investment Construction Concrete Joint Stock Company (DIC);

Table 5. Strength of concrete cement in the laboratory and on the site Compression strength, MPa Type of concrete 7 days 14 days 28 days

Tensile strength after 28 days, MPa

Conventional concrete cement 19.2 (19.6) 22.9 (23.3) 27.7 (28.1) 3.28 (3.35)

Concrete cement with 12% slag 17.4 (17.8) 21.4 (21.7) 26.1 (26.5) 2.89 (2.94)

Note: Result in the parenthesis in on the site.

3.2. Remark When applying to projects in fact, testing results in the laboratory is applied to reality. For

conventional concrete, compression strength in the laboratory is 1-3% different from in reality, tensile compression reaches 98% after 28 days when bending. For concrete with 12% slag, the compression strength in the laboratory is 1-3% different from in reality, tensile compression reaches 98% after 28 days when bending.

Figure 7. Some pictures of constructing the pilot section

by cement concrete containing mineral additive

3.3. Evaluation of economic efficiency To compare conventional cement concrete and 12% steel slag cement concrete for the

above works. When selecting 25Mpa cement concrete, cement content used in 1m3 concrete is compared

as follows: The price list is taken according to basic building price unit of Finance - Construction


Departments in October, 2011 of Ba Ria - Vung Tau province (the price doesn’t includes VAT - transported to the construction site).

In calculating expenses with depreciation of remaining value of production lines; revenue for support use of slag from steelmaking plants, the price of producing steel slag as a mineral additive is defined at approximately: 174,000 VND/ton.

Therefore, 1m3 cement concrete with steel slag is 62,654 VND cheaper than conventional concrete.

A road section with length of 430m, road surface width of 9.2m, road surface thickness of 15cm will require 593.4 m3 cement concrete. Use of cement with steel slag will be cheaper, saving 14,861,529 VND (forteen million eight hundred sixty one thousand five hundred twenty nine Vietnam dong), reducing 11% expense. However, the value doesn’t include benefit due to consuming large quantity of steel slag from steelmaking plants, harmful to the environment.

IV. CONCLUSION - When using steel slag as industrial waste as a mineral additive in cement concrete

components. It’s possible to partly replace and to reduce cement content, reduce hydration heat, to improve aggregate gradation, to enhance properties of cement concrete;

- For cement concrete using steel slag mineral particles, it’s required to replace part of cement according to percentage of 10%, 12% and 15%. Initial testing results for grades of 25, 30 and 35 MPa show that its mechanical properties meet technical requirements of cement concrete used for construction of motorway;

- On the other hand, when using steel slag slag particles as a mineral additive, it’s required to replace part of cement to minimize greenhouse effect and to make use of waste from steelmaking plants;

The study result shows that steel slag particles have characteristics to produce mineral additives meeting TCVN 6260-1997 on mineral additives in manufacturing cement and cement concrete.

In addition, use of steel slag from industrial waste remedies dumping and storage at concentrated waste treatment zone at Toc Tien commune, Ba Ria - Vung Tau province, which currently seriously pollutes the environment.

Economic value from cement concrete using steel slag as a mineral additive is also an issue of concern.

References [1]. Nguyễn Văn Tránh và Trần Vũ Minh Nhật - Đề tài Nghiên cứu dùng xỉ trong công nghiệp xản xuất xi măng Portland xỉ (RESEARCHING OF USING LASTFURNATED SLAG INPRODUCING SLAG-PORTLAND CEMENT INDUSTRY). [2]. PGS. TS Phạm Huy Khang - Tro bay và ứng dụng trong xây dựng đường ôtô và sân bay trong điều kiện Việt Nam. [3]. Thông tin Khoa học Kỹ thuật xi măng, số 1/2006 - Nghiên cứu hàm lượng xỉ lò cao tới độ bền sun phát của đá xi măng. [4]. Quyết định số 798/QĐ-TTg ngày 25-5-2011 của Thủ tướng Chính phủ về phê duyệt chương trình đầu tư xử lý chất thải rắn giai đoạn 2011-2020♦


MECHANISMS FOR INTEGRATION FEDERAL AND REGIONAL STRATEGY

FOR ENSURING ROAD TRAFFIC SAFETY IN RUSSIA

DR. ALEXANDER CHUBUKOV PROF. VALENTIN SILYANOV State Technical University - MADI

Summary: In the paper the effect of Federal target program on increment of road traffic safety in 2006-2012 in Russia is under discussion. The mechanisms for integration federal and regional strategies for ensuring road traffic safety are recommended.

Keywords: Federal target program; ensuring road traffic safety; road accidents.

Successful course of realization of the Federal target program « Increment of road traffic safety in 2006-2012», dynamics of change of target indicators of the program, as a whole positive reaction of a society to changes spent in this sphere, create premature optimistic estimations of the further development of a situation with road traffic safety in the regions of the Russian Federation.

Some results of realization of the Program (2004-2009) are the following:

• 8500 human lives are saved;

• Life of 562 children is saved;

• Preservation of indicators of 2004 would lead to death on roads of Russia of 13800 persons;

• The situation on roads of the Russian Federations has essentially improved;

• There is a full confidence of achievement of target indicators of the program.

Considering safety of traffic as one major, but not a unique indicator or the indicator of efficiency of functioning of an road transport complex of the Russian Federation, we should agree that its increase directly is connected with perfection of work of the basic components of transport system - technical, technological, economic, organizational-administrative, social, ecological, and also qualitative and quantitative interrelations and interdependence between


them. Proceeding from principles and rules of the system-target approach, we should recognize that now the main effect of realization of the Federal target program will reach basically at the expense of the decision of the most obvious, "simple" system problems, and the further preservation of positive tendencies of increase of safety of traffic will be defined in a greater degree by the decision of intersystem problems or, speaking more precisely problems of intersystem interaction and interdependence.

Examples of multisystem problems, which did not receive the decision:

• Organization and carrying out of the state vehicles technical inspection;

• Traffic management;

• Realization of projects within the limits of “Public-Private Partnership (PPP)”;

• Introduction in full the European protocol by the Russian insurers.

At the same time it is necessary to notice that the program realized till 2012 puts serious basic bases of the decision of similar problems, forms fundamental bases of the decision of interdepartmental and intersystem problems that under condition of maintenance of the further efficiency of the realized actions will provide preparation and performance of the second Federal target program “Increment of safety of traffic on 2013 – 2020”, focusing attention to internal motivation to observance of traffic regulations by all its participants, creation of a favorable climate on roads of Russia.

Outlined in first half of 2010 some delay of rates of positive change of some program indicators, indirectly confirms “exhaustible” influences of "simple" or monosystem actions on a traffic safety status.

Realization of mechanisms of integration of federal and regional programs, both within the limits of the new Federal target program, and with use of the resources put in other federal both regional programs and projects in sphere of public health services, formation, power, development of the industry, culture, agriculture can become one of directions of the further increase of safety of traffic.

Regional indicators of safety of traffic as result of interaction of a wide complex of factors and their essential dependence on level of social and economic development of region.

The detailed analysis of dynamics of change of indicators of safety of traffic on subjects of the Russian Federation indicates:


- Change of indicators of quantity of road accident, numbers of death and wounded men, social and transport safety, various kinds of road accident with participation of children, road accident because of drunk drivers differs from 2 to 11 times;

- In a number of regions negative dynamics of change of positive indicators of safety of traffic is observed;

- A number of regions show stably smaller in relation to average across Russia dynamics of change of indicators.

It is represented that the reason of similar statistical divergences can be caused as action subjective (organizational-administrative, professional, administrative), and objective (nature-climatic, economic, social) factors.

Subject of more detailed consideration will be the objective factors caused by features of given region and in many respects predetermining as level of safety of traffic in region, dynamics of its change, and set of the methods providing the most essential positive changes of safety of traffic. Such approach is represented justified, in connection with high expenses for realization of actions in sphere of safety of traffic taking into account a budgeted deficit, first of all in regions. It is necessary to notice that the formulated purposes of the Federal program are focused on achievement of Central European level that taking into account relative indicators of economic development of EU Member States and the Russian Federation it is represented enough ambitious and thus executable in practice by a problem. About it followed remember ardent, but to unsubstantiated critics of the program.

Among the general set of the indicators comprehensively characterizing development of subjects of the Russian Federation, it is possible to allocate two groups - direct and indirect, rendering, or able to influence indicators of safety of traffic in region.

The first group of indicators concern: motorization level, the indicators characterizing a condition of vehicle park, extent and high system density, quantity and number of inhabitants of settlements, relative charges of a high system, relative sales volume of alcoholic production through a retail network and some other.

To the second - the indicators characterizing cultural and educational qualifications of the population, employment, social activity, security services and development of system of public health services, a crime rate.

The total of indicators of both groups taken from the annual report of the State Statistical


Committee of the Russian Federation, makes nearby 30 and work on formation of such base is in an end stage.

Estimated, predesigns have shown that for various regions weightiness of those or other indicators of group fluctuates in essential limits and testifies to presence of interrelations, and with different quantitative interdependence. End of formation of system of significant indicators of both groups and carrying out of calculations by definition of their weightiness and interrelation with indicators of safety of traffic in concrete region, will allow to define priority spheres and directions of realization of a complex of actions in sphere of the road traffic management, economy, social development of the region, providing, taking into account a limit of financial assets, the maximum positive changes in safety of traffic.

The separate attention is deserved by a problem of realization of the detailed analysis of full statistics of regional road accidents taking into account all fixed parameters on accounting incidents. Experience of the countries of the European Union shows high effectiveness of operative actions in sphere of the traffic management and work of divisions similar by our traffic police in concrete directions, conditions and participants of traffic.

Integration and coordination of separate directions of Federal target programs, dot financing from regional budgets in actions and the projects providing the maximum return, taking into account regional features and priorities, are capable to provide synergistic effect in the field of safety of traffic.

At the same time at sessions of the Governmental Road Traffic Safety Commission the question of principle about necessity of investment by function of the practical organization of traffic of uniform body, jurisdiction and which powers would extend on a roads and streets network of all without an exception of subjects was repeatedly discussed: a federal road network, an road-street system of subjects of federation, municipal and rural formations. Occurrence of barriers on "borders", and the most important thing of their consequence have brightly been shown in the summer of 2010 at overpass repair on 24 kilometer of the Moscow-Leningrad highway near the International Airport "Sheremetyevo".

In the conclusion it is necessary to notice that the high lath of indicators of safety of the traffic, set by the Federal target program, focused on the European indicators of safety, should be confirmed including by corresponding level of the "European" capital investments in this sphere, including material compensation of employees of all structures providing safety of traffic on roads of Russia.


BOARD OF EDITORS - IN - CHIEF

Prof. V. Prikhodko; Prof.V. Silyanov; Prof. Wanming Zhai; Assoc.Prof. Nguyen Van Vinh

EDITORIAL COUNCIL

MADI’s Editorial SWJTU’s Editorial UTC’s Editorial

Prof. A. Buslaev Prof. Dr. Zhao Yong Assoc.Prof. Dr.Tran Dac Su

Prof. A. Chubukov Prof. Dr. Zhai Wanming Assoc.Prof. Dr.Nguyen Van Vinh

Prof. I. Demyanushko Prof. Dr. Qiu Yanjun Assoc.Prof. Dr.Nguyen Duy Viet

Prof. I. Fedorov Prof. Dr. Li Qiao Assoc.Prof. Dr.Nguyen Ngoc Long

Prof. V. Gerami Prof. Dr. Gao Shibin Assoc.Prof. Dr. Tran Tuan Hiep

Prof. A. Ivanov Prof. Dr. Peng Qiyuan Assoc.Prof. Dr. Nguyen Van Long

Prof. L. Makovskyi Prof. Dr. Pan Wei Dr. Nguyen Quynh Sang

Prof. A. Nikolayev Prof. Dr. Li Fu Prof. Dr. Do Duc Tuan

Prof. V. Nosov Prof. Huang Nan Assoc.Prof. Dr. Bui Xuan Cay

Prof. P. Pospelov Prof. Dr. Liu Xueyi Dr. Le Hai Ha

Prof. A. Rementzov Prof. Dr. Zheng Kaifeng Assoc.Prof. Dr. Nguyen Van Bang

Prof.. M. Shatrov Prof. Dr. Zhang Jin Assoc.Prof. Dr. Do Viet Dung

Prof. Yu. Trofimenko Prof. Dr. Liu Dan Assoc.Prof. Dr. Le Hung Lan

Prof. M. Ulitskyi Prof. Dr. Yang Yiren Assoc.Prof. Dr. Tu Sy Sua

Prof. V. Vlasov Prof. Liu Bin Prof. Dr. Sc Nguyen Huu Ha

Prof. V. Zhurakovskyi Prof. Dr. Song Jirong Assoc.Prof. Dr. Le Hong Lan

Prof. V. Zorin Dr. Nguyen Tai Quang

SECRETARY SECTION

MSc. V. Vinogradova Assoc. Prof. Lan Junsi Dr. Nguyen Quynh Sang

MSc. V. Lipskaya

of transportation - Главнаяlib.madi.ru/fel/fel1/fel12p004.pdf · international cooperation...

Documents