modelling and simulation of tramway transportation...
TRANSCRIPT
Research ArticleModelling and Simulation of Tramway Transportation Systems
A Capasso1 M Ceraolo2 R Lamedica1 G Lutzemberger 2 and A Ruvio1
1Sapienza University of Rome Rome Italy2Department of Energy Systems Territory and Constructions Engineering University of Pisa Largo L Lazzarino n 1 Pisa Italy
Correspondence should be addressed to G Lutzemberger lutzembergerdseaunipiit
Received 18 September 2018 Revised 24 December 2018 Accepted 3 January 2019 Published 3 February 2019
Guest Editor Michela Longo
Copyright copy 2019 A Capasso et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited
Electrified guided vehicles typically face routes having a large number of acceleration and braking phasesThe braking energy sincethe feeding line presents nonreversible electrical feeding substations can be recovered in the presence of other nearby vehiclesTo improve braking energy recovery one or more storage systems can be positioned along the track Analysis of effectivenessfor the considered solution requires time-domain simulation models to be created through suitable simulation general-purposelanguages or specialised languagessoftware In this paper three different tools for the considered existing tramway were developedand the main examined characteristics have been compared to each other Then analysis of output results was also performeddemonstrating the real cost-effectiveness of introducing one storage device on the considered tramline in operation
1 Introduction
Electrified guided vehicles such as trams typically face routeswith a large number of acceleration and braking phasesIn this regard only a fraction of the initial kinetic energycan be partially recovered [1ndash4] In particular if the feedingline presents nonreversible electrical feeding substationsbraking energy can partially get back only in the presence ofother nearby vehicles capable of adsorbing traction energyTo improve the possibility of recovering energy stationarystorage systems along the track can be installed in order toadsorb the energy during braking when no other trains areable to receive it
Analysis of the cost-effectiveness of the proposed solutioninvolves the development of time-domain simulation toolsObviously solicitations on the storage system are influencedby timetable of the trams positioning of substations etcTypically the storage is installed through the interposition ofa DCDC converter in order to limit current within safetylimits or to avoid battery State-of-Charge drift to maintainunaltered its capability to adsorb or deliver energy
In order to make time-domain simulations it is pos-sible to create models by utilisation of general-purposeprogramming languages ie FORTRAN or C or specialisedlanguages or software ie Modelica or Matlab-Simscaperespectively
As example of modelling and simulation in railway ap-plications modelling of short-circuit protections or DCelectrical railway systems to simulate stray currents and touchvoltages has been already developed in [5 6]
In this paper simulation tools aimed at correctly rep-resenting the railway system including the electric powersupply the storage systems and the vehicles moving alongthe rails have been widely developed In particular threesimulation models were developed The first tool has beenrealised in FORTRAN It has been made a long time ago andused in many applications [7 8] Several experimental testsin order to make its validation were also performed in thepast [9]
On the other hand Dymola [10] a commercial tool basedon the open-source Modelica language [11] represents arecent solution having many advantages in terms of flexibil-ity simulation efficiency and man-machine interface as alsodemonstrated in several past works on railway systems by thesame authors [12 13]
In the last years many other commercial tools were alsodeveloped asMatlab-Simscape [14]This last one is not basedon an open-source environment but it is a relatively newtoolbox of one of the most considered software tools foracademic and company uses Indeed it is of some interestto compare the different tools as started in [15] also byextending the comparison to the new ones More precisely
HindawiJournal of Advanced TransportationVolume 2019 Article ID 4076865 8 pageshttpsdoiorg10115520194076865
2 Journal of Advanced Transportation
the newest tools were firstly confirmed on results achieved bythe first one in order to ensure the strength of the obtainedresults Then powerful and flexibility of each one have beenaccurately verified
Therefore after mutually validating the three tools underconsideration a technical-economic analysis for a tramwayline in operation has been performed In this way energyconsumption in realistic traffic conditions was taken intoaccount and energy saving due the installation of a storagedevice accurately calculated Finally payback time of theinvestment was evaluated thus demonstrating the cost-effectiveness of the considered solution
2 The Simulation Tools
As said in order to evaluate the amount of the brakingrecoverable energy it is mandatory to develop a simulationtool capable of correctly simulating the feeding network andthe vehicles dynamic Then different running phases andfrequency of the trains have to be considered Followingmodelling criteria were implemented
21 FORTRAN Language Based Model FORTRAN languageis a consolidate code to develop electrical model to repre-sented power network For DC railway application calcula-tion code Train-sim consisting of two computational tools ispresented in [7 8]
The first one allows calculating all the electromechanicalcharacteristics and performance of trains on a specific railwayline Based on altimetry profile of the track and rolling stockfeatures it is possible to carry out the train performancedue to motion stages Dynamic and kinematic profiles areobtained It is possible to set also various traffic scenariosThetool also calculates the amount of recoverable energy at eachbraking phase
The second one makes the DC electrified network loadflow calculation It builds an equivalent electrical network ateach step based on the tram positions along the route ESSsvehicles and parallel points are the electrical nodes of theconsidered equivalent network
The calculation code due to the model reported belowpermits the calculation of the electrical parameter of the net-work considering also the energy recovered during brakingphase The models are as follows
(i) Electrical substation (ESS) Figure 1 highlights theequivalent circuit to represent the voltage-currentcharacteristic of a typical electrical substation Whenthe substation is not connected (zone 2) the node Ais settled with P = 0 During supply or regenerationmode V = V
0condition (zone 1 no-load output
voltage) orV =119881rec (zone 3 network recovery voltage)is imposed at the node through resistance R
3or R1
to simulate different voltage drop as shown in zone 3and zone 1
(ii) Trams in traction The vehicles are represented asa constant power load However if the currentexceeds the maximum allowed value during opera-tion the constraint I=119868max (constant-current opera-tion) is imposed during the iterative calculation of
V
I
1
2
3
A
6L=
6I
6I
21
23
Network - Line
Figure 1 Voltage-current characteristic of an electrical substation(ESS)
the nonlinear equations until the voltage at panto-graph decreases Standard EN 50388 [16] imposedVlowast=08119881max so the model reproduces the V-P char-acteristic in agreement with this value
(iii) Trams in braking Braking energy recovering dependson the receptivity of the system If the system allowsit the traction line can accept all the power generatedduring braking phase according to the solution of thesystem equations If the voltage increases up to 119881maxthe recoverable energy transferred to the catenaryis lower Then the distance among the recoverableenergy and the energy effectively recovered gives anidea of the energy dissipated on braking rheostatIn this regard a mixed rheostat-regenerative braking(shown as dash-dotted line in Figure 2) can besimulated Figure 2 shows how the model works dueto this braking technique in particular the currentdissipated on the braking rheostat increases andconsequently the current entered into the tractionline decreases up to being zeroed when the voltage119881max is reached Typically V rsquo is in the range 09divide095119881max However a good strategy to avoid the voltagedecrease just explained is following the solid linereported in Figure 2 where V rsquo=119881max
Finally simple laws implemented for traction (1) andbraking (2) according to the above points (see Figure 2) areas follows
119881min lt 119881 lt 119881lowast 997904rArr 119868 = cost le 119868max
119881lowast lt 119881 lt 119881max 997904rArr 119875 = cost(1)
119881 lt 119881max 997904rArr 119875 = cost
119881 = 119881max 997904rArr 119875 = 0(2)
Journal of Advanced Transportation 3
V node Cond NodeTraction
Braking
P
V
6GCH 6lowast 6 6GR
Vlowast = 08 Vmax
6GCHlt6lt6lowast
6lowastlt6lt6GR
6lt6GR
6=6GR
I=costle)GR
P=cost
P=costP=0
Figure 2 Voltage-power characteristic during traction and brakingstages
For each electrical substation the software calculatedelectrical parameters about power current etc including theamount of recoverable energy during braking and the voltagealong the line
The procedure for determining the matrix coefficients ofthe admittance constituting the traction line refers to electri-cal networks in permanent sinusoidal regime and thereforeto complex admittance and impedanceThe software dealingwithDCnetworks considers all themagnitudes to be real theadmittance as conductance and the impedance as resistance
Regarding numerical solving methods Newton-Raphsonor Gauss-Seidel method is typically applied Some variantsof Newton-Raphson method particularly suitable for small-and medium-size networks can be also considered in orderto improve conditions of convergence and to reduce numberof iterations [17]
22 Modelica and Matlab-Simscape Based Model As saidthe simulation tool realised in Dymola is based on Mod-elica language [11] It is a single simulator performing thesame functionalities of the two previous ones written inFORTRAN to the advantage of accuracy and flexibility aswill be analysed later As anticipated Modelica language iscyber-physical and allows simulating complex systems iehaving mechanical electrical thermal control subsystemsFollowing the proposed approach the ESSs were simulatedthrough lumped components linked in a graphical wayas visible in Figure 3 As this study shows only the DCcomponent is of interest to evaluate effects of harmonics eachsubstation has been modelled through its well-known DCequivalent Naturally the system is time-variant In fact thecontact-line configuration varies with time since the trainposition varies In fact resistor values change depending onthe train position according to the following expressions
1198771= (1 minus 120575) 119877
119905119900119905
1198772= 120575119877119905119900119905
120575 =(1198752minus 119909)
11987112
(3)
whereR1is the left side line resistance R
2is the right side line
resistance and 120575 is the ratio of the distance between the trainposition (x) and ESS2 position (P
2) and the distance between
ESS1 and ESS2 substations (L12)
In addition the electrical feeding substations in operationare subjected to variation when a train moves from asection to another along the track This difficulty can beeasily addressed in Modelica considering the possibility ofchanging the system equations after some events happen[12 13]
Finally modelling of trains requires modelling of theelectric drive resistance forces and the driverrsquos behaviourElectric drive ismodelled as a system able to produce the trac-tive force as required by the driver within the allowed forceand power limits generating some power losses expressed asa function of the mechanical speed and the required forceThen each train must avoid feeding power to the catenarywhen this would cause the line voltage to become too largeand a controller of the DC power must be implementedA much more sophisticated control strategy has been usedhaving feedback on the instantaneous pantograph voltage andmodulating the braking power conveyed along the catenaryin order to avoid reaching instantaneously the upper allowedlimit Resistance to movement has been modelled using theformula including aerodynamic drag and rolling resistance
119877 = 119898119892 (119891119903cos 120572 + sin 120572) + 119860119881 + 1198611198812 (4)
whereR is the global resistance force acting against the vehiclemovement 120572 is the angle between the track and the horizon-tal plane m is the vehicle mass 119891
119903is the rolling resistance
coefficient A and B are empirical positive numbers takinginto account the lateral and front aerodynamic resistance andV is the train speed About the driver it can be representedsimply by a proportional controller in which the referencetractive force is proportional to the error between the actualand the reference speed
Further details regarding different submodel and controllogic in particular the blending strategy depicted in Figure 2are widely described also in [15]
In parallel with object-oriented interface the user candirectly insert physical or control equations These last arewritten exactly as in textbooks Starting from individualsubsystems description Modelica-based tool automaticallyperforms many operations first the identification of a set ofdifferential algebraic equations (DAEs) representing the sys-tem under study and then after some additional operationsto simplify the set of equations [15] the conversion in a Cor C++ language compilation to make the final simulationexecutable This task of automatic creation of simulationexecutable requires just a few seconds So changing thesystem to create a new executable (for instance to have moretrains or a different number of ESSs) will require just arepetition of this automatic process Once the executable hasbeen created it can run several times while changing someparameters from a run to another [15]
Last simulator is realised in Matlab-Simscape [14] andentirely developed starting from the tool realised inModelicain this regard the model architecture and equations of the
4 Journal of Advanced Transportation
Train 3
traction circuitequivalent resistors
+ +
Train 2Train 1
ESS1 ESS2
EKEK
RK RK
R1(t) P1u R2(t) P2uR3(t) R4(t)
P1l P2l
Figure 3 Graphical representation of the electrical feeding system
submodels are the same already implemented in ModelicaThus lumped components have been simply rebuilt usingMatlab-Simscape libraries always according to Figure 3 andconsidering for their control part the well-known Matlab-Simulink control libraries
23 Comparisons The three tools were tested by analysingtheir flexibility simulation efficiency and man-machineinterface
Flexibility This aspect was identified in terms of capabilityto model new case studies The Modelica-based tool is ableto rapidly modify the existing models through its object-oriented interface simply by changing or connecting newelements Indeed it is possible to increase or reduce thenumber of the trains running the number and positioningof the electrical feeding substations the eventual storagesystems located along the track the auxiliary adsorption etcIndeed expandability of the model is easily guaranteed alsoby the fact that theModelica-based tool is a unique simulatorunlike those realised in FORTRAN Additionally creation ofthe simulation executable requires just few seconds Thuschanging the system simply requires the rapid repletion ofthis process These characteristics may be retrieved also inMatlab-Simscape although the graphical interface is lessclear and intuitive as only partially object-oriented due tothe need for additional Matlab-Simulink control blocks Onthe other hand the FORTRAN language based model iscompletely different requiring to be fully recompiled at everysmall change [18]
Simulation Efficiency This was evaluated by speed and mem-ory requirements It must be said that the FORTRAN numer-ical solver may be optimised for the considered case studyInstead Modelica and Matlab-Simscape utilise standardnumerical solvers not specifically optimised on the singlecase study The comparison was executed having one systemcharacterised by a single electrical feeding substation and twotrams on the track As already described in [15] at equalnumber of samples and time length simulation theModelica-based tool engaged 115 s requiring 6 MB of memory Thisis nearly the same for Matlab-Simscape which employsabout 20 s to perform a simulation of 1000 s with memoryoccupation of 11 MB In order to consider a proper time forcomplete simulation for FORTRAN calculation code it isneeded to consider the time duration to move between thetwo tools In the case of the considered simulation (ie 1000s) it engaged about 98 s and 20MBofmemory requirements
0100020003000400050006000
0 200 400 600 800 1000
Dist
ance
(m)
time (s)
Figure 4 Pattern profile
Man-Machine Interface The tool developed in Modelicaallows the possibility of making changes in different waysFirst equations can be easily changed by text interface Onthe other hand model structure and related parameters maybe changed by graphical interface [13] This is nearly thesame as utilisation of the Matlab-Simscape tool Regardingthe FORTRAN based model Visual-Basic language is usedto manage input and output data in Train-sim software thesupport of the user friendly macros implemented permitsmodifying parameters in the software although much moreslowly than the others
In conclusion the Modelica-based and Matlab-Simscapetools require high memory requirements but also guaranteemuch more flexibility The FORTRAN based tool developedmany years ago is very useful because it is able to producebenchmark results to be used asmain reference On the otherhand the other two tools allow fast creation of models whosesimulation results need to be carefully verified
3 Case Study
31 Validation The tools have been tested having as referencecase study an existing tramway in Rome The path length isabout 57 km as noticeable from Figure 4 having one singleelectrical feeding substation two trains are on rails one foreach track The main input model parameters are listed inTable 1
The simulation results were evaluated in terms of energyand power flows of the considered tramway In the first exam-ined condition the trams make use of on-board resistors todissipate all the braking energy In the second scenario theysend braking energy on the catenary until the voltage doesnot reach themaximumadmitted value fixed at 800V Finallytrams send braking energy as before but with one storagesystem installed about halfway along the tramway
Journal of Advanced Transportation 5
Table 1 Input model parameters of the system under study
ESS parametersNo load voltage (V) 1680R0(Ω) 013
Number of ESSs 1ESS position (km) 0
Line parametersMax line voltage (V) 1800Nominal line voltage (V) 1650Min line voltage (V) 1100Number of line trunks 6
Line Resistance (Ωkm) 010 010 007007 007 007
Tram parametersFull mass (t) 92Auxiliary power adsorption (kW) 40ED max power (kW) 1242ED max traction force (kN) 96
Track parametersNumber of trams 2Number of stops 14Average distance between stops (km) 04Max speed (ms) 14Track length (km) 54
First simulations were performed without consideringbraking energy recovery As said model parameters wereslightly updated in order tomatch the results provided by thenew tools with the oldest one Figure 5 shows the plot resultsof one train driving on a portion of the considered route Inthis case the tram is not able to send its braking energy alongthe catenary As observable plot results are equivalent amongthe three tools respectively developed in Modelica Matlab-Simscape and FORTRAN
Then simulations have considered the possibility ofrecovering the braking energy In this case parameters actingon control voltage at pantograph have been tuned to performmodulation of the inlet power without overcoming voltagelimits Figure 6 shows plot results of one train under a portionof the considered profile
As noted different blending strategies according to Fig-ure 2 have been implemented Naturally these differenceshave an impact on the total energy absorbed from thenetwork The blending strategy having Vrsquo=09divide095119881max (seeFigure 2) implemented in FORTRAN results in a reductionof the energy recovered through the pantograph respectivelyof 13 and 2 with respect to the dynamical strategyused in Modelica and Matlab-Simscape that allows the fullpowertrain power to be recovered up to 119881max (ie V rsquo=119881max)Then in terms of total energy adsorption from the ESSincrement is of 3 and 06 respectively
Naturally braking energy recovery can be enhancedthrough energy storage systems installed along the route
Table 2 Lithium battery main characteristics
Nominal energy (MWh) 033Nominal capacity (Ah) 200Nominal voltage (V) 1650Number of cells in series 446Max allowed current (A) 2000Charging-discharging efficiency 09
Table 3 ESS energy consumption
ESS working day daily energy (MWh) 158ESS holiday daily energy (kWh) 111ESS Annual energy (MWh) 5386
In this way one storage system has been introduced withthe main aim of validating the tools under test also in thenew considered system configuration The lithium battery ispositioned about halfway along the tramway (ie about 38km from the terminal) whose characteristics are in Table 2
Relation among the nominal energy and nominal capac-ity is given by
119864119899= 119899119904119881119899119888119862119899 (5)
where 119899119904is the number of cells in series 119864
119899is the nominal
energy 119881119899119888is the nominal cell voltage and 119862
119899is the nominal
capacity Number of cells and nominal capacity were selectedin order to exactly reproduce through the newest tools thebattery SOC (State-of-Charge) and the power profile withrespect to the FORTRAN based one Results are shown inFigure 7
Installation of one storage system can significantly reducethe energy delivered by the electrical feeding substations(ESSs) The considered simplified case study with two trainsobviously cannot correctly evaluate the cost-effectivenessof the considered solution Thus the full analysis will bedescribed in the next section following the same approachalready considered by the authors in [12 13]
32 Energy Saving Evaluation Energy saving evaluationhas been performed by considering an experimental mea-surement campaign carried on by the authors Results aresummarized in Table 3 where the daily energy delivered fromthe electrical substation (ESS) has been measured duringone typical school working day and one typical holidayday Starting from that evaluation of the annual electricalenergy demand has been performed simply by consideringthe number of working days and holidays over the year givenby the tramline operator
Therefore the Modelica-based tool presented before hasbeen used to exactly reproduce the energy consumptionshown in Table 3 To do that different numbers of operatingtrams along the day have been considered from timetablerespectively 24 during peak hours typically concentrated inthe middle part of the day and 12 in low load hours mainlyin the early morning and in the late evening Real numberof trams in operation may be slightly different However the
6 Journal of Advanced Transportation
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
Tram
1 V
olta
ge (V
)
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
times106
Figure 5 Plot results trams without braking energy recovery
daily distribution has been chosen to exactly reproduce themeasured consumption
After simulating exactly the same level of the measuredenergy demand ie obtaining the daily consumption shownin Table 3 simulations have been repeated by including thestorage system described in Table 2 The delivered energy bythe ESS was then compared with the previous one calculatedin absence of the storage system The reduction in annualenergy consumption was about 152 moving from 5386MWh up to 4570MWh
The cost-effectiveness of the proposed solution has beeninvestigated by considering the initial cash outlay due theintroduction of the storage system with respect to theannual return of the investment due to the above-mentionedelectrical energy saving The initial cash outlay due to thestorage system and its balance of plant has been calculatedby considering a value of 500 eurokWh including cells BMSand battery packaging As discussed in detail in [19] thepresence of a DCDC converter may be required For thisa fixed cost of 60 keuro in analogue way to the experiencepresented in [19] has been used Finally the current industrialuser price of energy in Italy was evaluated considering anaverage value of 150 euroMWh In terms of maintenancecosts for the reasons already explained in [19] the storagehas been considered able to cover the whole life of theplant
Table 4 Economic benefit analysis
Storage system cost (keuro) 2261Annual energy saving (keuro) 1225NPV (keuro) 7382PBT (y) 2
Main objective of the analysis was related to the evalua-tion of the net present value (NPV) and of the payback time(PBT) for a whole life of ten years and an interest rate of 4Results are shown in Table 4
The results show that installation of a stationary storagesystem may guarantee a payback time within just two yearThese numbers are so favourable not to be affected by anypossible storage substitution during the plant life Indeed thecost-effectiveness of the proposed solution has been clearlydemonstrated
4 Conclusions
This paper has demonstrated how innovative languages andsoftware allow rapid creation of numerical models havingelectrical mechanical and control parts to be simulated
The Modelica-based model was realised through thecommercial Dymola tool However it could run inside any
Journal of Advanced Transportation 7
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
2000
Tram
1 V
olta
ge (V
)
Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
times106
Figure 6 Plot results trams with braking energy recovery
0 100 200 300 400 500 600 700 800 900 1000Time (s)
04
045
05
055
RESS
SO
C (p
u)
Simscape Dymola Fortran
Simscape Dymola Fortran
0 100 200 300 400 500 600 700 800 900 1000Time (s)
minus1
0
1
2
RESS
pow
er (W
)
times106
Figure 7 Plot results system equipped with storage
8 Journal of Advanced Transportation
other Modelica-compliant tool They proved to be fast withrelatively small memory occupation Nearly the same can besaid about Matlab-Simscape although characterised by lessflexibility due to the fact that it is not inspired by an open-source language platform The quality has been verified bycomparing results with those obtained using a well validatedFORTRAN based simulator
With reference to an existing case study the cost-effectiveness due to the utilisation of stationary storagesystem to enhance braking energy recovery has been clearlydemonstrated since on a high-traffic tramway with highnumber of stops payback time has been reached within onlytwo years Naturally it is of great importance in order to cor-rectly achieve energy saving for the considered application topreliminary calibrate the considered tool on actual electricalenergy consumption experimentally measured
As future direction of this work it is also possible toconsider the extension of the considered methodology toother tramlines in operation in order to investigate thepotential cost-effectiveness of similar or different energysaving solutions
Data Availability
The data used to support the findings of this study areincluded within the article
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] A Gonzalez-Gil R Palacin P Batty and J P Powell ldquoA systemsapproach to reduce urban rail energy consumptionrdquo EnergyConversion and Management vol 80 pp 509ndash524 2014
[2] A Gonzalez-Gil R Palacin and P Batty ldquoSustainable urban railsystems Strategies and technologies for optimalmanagement ofregenerative braking energyrdquo Energy Conversion and Manage-ment vol 75 pp 374ndash388 2013
[3] M C Falvo R Lamedica and A Ruvio ldquoEnergy storage appli-cation in trolley-buses lines for a sustainable urban mobilityrdquoin Proceedings of the 2012 Electrical Systems for Aircraft Railwayand Ship Propulsion ESARS 2012 Italy October 2012
[4] M C Falvo R Lamedica and A Ruvio ldquoAn environmentalsustainable transport system A trolley-buses Line for Cosenzacityrdquo in Proceedings of the 21st International Symposium onPower Electronics Electrical Drives Automation and MotionSPEEDAM 2012 pp 1479ndash1485 Italy June 2012
[5] M Berger C Lavertu I Kocar and J Mahseredjian ldquoProposalof a time-domain platform for short-circuit protection analysisin rapid transit train DC auxiliary systemsrdquo IEEE Transactionson Industry Applications vol 52 no 6 pp 5295ndash5304 2016
[6] A Ibrahem A Elrayyah Y Sozer and J A De Abreu-GarcialdquoDC railway system emulator for stray current and touchvoltage predictionrdquo IEEE Transactions on Industry Applicationsvol 53 no 1 pp 439ndash446 2017
[7] A Capasso R Lamedica andC Penna ldquoEnergy regeneration intransportation systemmethodologies for powernetworks simu-lationrdquo in Proceedings of the 4th IFACIFIPIFORS international
conference on control in transportation system Baden GermanyApril 1983
[8] A Capasso G Giannini R Lamedica and A Ruvio ldquoEco-friendly urban transport systems Comparison between energydemands of the trolleybus and tram systemsrdquo Ingegneria Fer-roviaria vol 69 no 4 pp 329ndash347 2014
[9] A Adinolfi R Lamedica C Modesto A Prudenzi and SVimercati ldquoExperimental assessment of energy saving due totrains regenerative braking in an electrified subway linerdquo IEEETransactions on Power Delivery vol 13 no 4 pp 1536ndash15411998
[10] Dymola ldquoDynamic Modeling Laboratoryrdquo documentationavailable from httpwww3dscomproductsservicescatiaproductsdymola
[11] P Fritzson Introduction toModeling and Simulation of Technicaland Physical Systems with Modelica Wiley-IEEE Press 2011
[12] M Ceraolo R Giglioli G Lutzemberger M Conte and MPasquali ldquoUse of electrochemical storage in substations toenhance energy and cost efficiency of tramwaysrdquo in Proceedingsof the 2013 AEIT Annual Conference Innovation and Scientificand Technical Culture for Development AEIT 2013 MondelloItaly October 2013
[13] S Barsali P Bolognesi M Ceraolo M Funaioli and G Lu-tzembergerCyber-Physical Modelling of Railroad Vehicle SystemusingModelica Simulation Language vol 104 Railways AjaccioCorsica France 2014
[14] ldquoMatlab-Simscaperdquo Official site httpswwwmathworkscomproductssimscapehtml
[15] A Capasso R Lamedica A Ruvio M Ceraolo and GLutzemberger ldquoModelling and simulation of electric urbantransportation systems with energy storagerdquo in Proceedings ofthe 16th International Conference on Environment and ElectricalEngineering (EEEIC rsquo16) Florence Italy June 2016
[16] ldquoCEI EN 50388 Railway applications - Power supply and rollingstockrdquo 2012
[17] M E El-Hawary and O K Wellon ldquoThe alpha-modified quasi-second order Newton-Raphson method for load flow solutionsin rectangular formrdquo IEEE Transactions on Power Apparatusand Systems vol 101 no 4 pp 854ndash866 1982
[18] A Capasso and R Lamedica Simulation of The Operation ofElectric Traction Networks CNR - Progetto finalizzato trasportiBologna Italy 1984
[19] M Ceraolo R Giglioli G Lutzemberger and A BechinildquoCost effective storage for energy saving in feeding systems oftramwaysrdquo in Proceedings of the 2014 IEEE International ElectricVehicle Conference IEVC 2014 IEEE Florence Italy December2014
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
2 Journal of Advanced Transportation
the newest tools were firstly confirmed on results achieved bythe first one in order to ensure the strength of the obtainedresults Then powerful and flexibility of each one have beenaccurately verified
Therefore after mutually validating the three tools underconsideration a technical-economic analysis for a tramwayline in operation has been performed In this way energyconsumption in realistic traffic conditions was taken intoaccount and energy saving due the installation of a storagedevice accurately calculated Finally payback time of theinvestment was evaluated thus demonstrating the cost-effectiveness of the considered solution
2 The Simulation Tools
As said in order to evaluate the amount of the brakingrecoverable energy it is mandatory to develop a simulationtool capable of correctly simulating the feeding network andthe vehicles dynamic Then different running phases andfrequency of the trains have to be considered Followingmodelling criteria were implemented
21 FORTRAN Language Based Model FORTRAN languageis a consolidate code to develop electrical model to repre-sented power network For DC railway application calcula-tion code Train-sim consisting of two computational tools ispresented in [7 8]
The first one allows calculating all the electromechanicalcharacteristics and performance of trains on a specific railwayline Based on altimetry profile of the track and rolling stockfeatures it is possible to carry out the train performancedue to motion stages Dynamic and kinematic profiles areobtained It is possible to set also various traffic scenariosThetool also calculates the amount of recoverable energy at eachbraking phase
The second one makes the DC electrified network loadflow calculation It builds an equivalent electrical network ateach step based on the tram positions along the route ESSsvehicles and parallel points are the electrical nodes of theconsidered equivalent network
The calculation code due to the model reported belowpermits the calculation of the electrical parameter of the net-work considering also the energy recovered during brakingphase The models are as follows
(i) Electrical substation (ESS) Figure 1 highlights theequivalent circuit to represent the voltage-currentcharacteristic of a typical electrical substation Whenthe substation is not connected (zone 2) the node Ais settled with P = 0 During supply or regenerationmode V = V
0condition (zone 1 no-load output
voltage) orV =119881rec (zone 3 network recovery voltage)is imposed at the node through resistance R
3or R1
to simulate different voltage drop as shown in zone 3and zone 1
(ii) Trams in traction The vehicles are represented asa constant power load However if the currentexceeds the maximum allowed value during opera-tion the constraint I=119868max (constant-current opera-tion) is imposed during the iterative calculation of
V
I
1
2
3
A
6L=
6I
6I
21
23
Network - Line
Figure 1 Voltage-current characteristic of an electrical substation(ESS)
the nonlinear equations until the voltage at panto-graph decreases Standard EN 50388 [16] imposedVlowast=08119881max so the model reproduces the V-P char-acteristic in agreement with this value
(iii) Trams in braking Braking energy recovering dependson the receptivity of the system If the system allowsit the traction line can accept all the power generatedduring braking phase according to the solution of thesystem equations If the voltage increases up to 119881maxthe recoverable energy transferred to the catenaryis lower Then the distance among the recoverableenergy and the energy effectively recovered gives anidea of the energy dissipated on braking rheostatIn this regard a mixed rheostat-regenerative braking(shown as dash-dotted line in Figure 2) can besimulated Figure 2 shows how the model works dueto this braking technique in particular the currentdissipated on the braking rheostat increases andconsequently the current entered into the tractionline decreases up to being zeroed when the voltage119881max is reached Typically V rsquo is in the range 09divide095119881max However a good strategy to avoid the voltagedecrease just explained is following the solid linereported in Figure 2 where V rsquo=119881max
Finally simple laws implemented for traction (1) andbraking (2) according to the above points (see Figure 2) areas follows
119881min lt 119881 lt 119881lowast 997904rArr 119868 = cost le 119868max
119881lowast lt 119881 lt 119881max 997904rArr 119875 = cost(1)
119881 lt 119881max 997904rArr 119875 = cost
119881 = 119881max 997904rArr 119875 = 0(2)
Journal of Advanced Transportation 3
V node Cond NodeTraction
Braking
P
V
6GCH 6lowast 6 6GR
Vlowast = 08 Vmax
6GCHlt6lt6lowast
6lowastlt6lt6GR
6lt6GR
6=6GR
I=costle)GR
P=cost
P=costP=0
Figure 2 Voltage-power characteristic during traction and brakingstages
For each electrical substation the software calculatedelectrical parameters about power current etc including theamount of recoverable energy during braking and the voltagealong the line
The procedure for determining the matrix coefficients ofthe admittance constituting the traction line refers to electri-cal networks in permanent sinusoidal regime and thereforeto complex admittance and impedanceThe software dealingwithDCnetworks considers all themagnitudes to be real theadmittance as conductance and the impedance as resistance
Regarding numerical solving methods Newton-Raphsonor Gauss-Seidel method is typically applied Some variantsof Newton-Raphson method particularly suitable for small-and medium-size networks can be also considered in orderto improve conditions of convergence and to reduce numberof iterations [17]
22 Modelica and Matlab-Simscape Based Model As saidthe simulation tool realised in Dymola is based on Mod-elica language [11] It is a single simulator performing thesame functionalities of the two previous ones written inFORTRAN to the advantage of accuracy and flexibility aswill be analysed later As anticipated Modelica language iscyber-physical and allows simulating complex systems iehaving mechanical electrical thermal control subsystemsFollowing the proposed approach the ESSs were simulatedthrough lumped components linked in a graphical wayas visible in Figure 3 As this study shows only the DCcomponent is of interest to evaluate effects of harmonics eachsubstation has been modelled through its well-known DCequivalent Naturally the system is time-variant In fact thecontact-line configuration varies with time since the trainposition varies In fact resistor values change depending onthe train position according to the following expressions
1198771= (1 minus 120575) 119877
119905119900119905
1198772= 120575119877119905119900119905
120575 =(1198752minus 119909)
11987112
(3)
whereR1is the left side line resistance R
2is the right side line
resistance and 120575 is the ratio of the distance between the trainposition (x) and ESS2 position (P
2) and the distance between
ESS1 and ESS2 substations (L12)
In addition the electrical feeding substations in operationare subjected to variation when a train moves from asection to another along the track This difficulty can beeasily addressed in Modelica considering the possibility ofchanging the system equations after some events happen[12 13]
Finally modelling of trains requires modelling of theelectric drive resistance forces and the driverrsquos behaviourElectric drive ismodelled as a system able to produce the trac-tive force as required by the driver within the allowed forceand power limits generating some power losses expressed asa function of the mechanical speed and the required forceThen each train must avoid feeding power to the catenarywhen this would cause the line voltage to become too largeand a controller of the DC power must be implementedA much more sophisticated control strategy has been usedhaving feedback on the instantaneous pantograph voltage andmodulating the braking power conveyed along the catenaryin order to avoid reaching instantaneously the upper allowedlimit Resistance to movement has been modelled using theformula including aerodynamic drag and rolling resistance
119877 = 119898119892 (119891119903cos 120572 + sin 120572) + 119860119881 + 1198611198812 (4)
whereR is the global resistance force acting against the vehiclemovement 120572 is the angle between the track and the horizon-tal plane m is the vehicle mass 119891
119903is the rolling resistance
coefficient A and B are empirical positive numbers takinginto account the lateral and front aerodynamic resistance andV is the train speed About the driver it can be representedsimply by a proportional controller in which the referencetractive force is proportional to the error between the actualand the reference speed
Further details regarding different submodel and controllogic in particular the blending strategy depicted in Figure 2are widely described also in [15]
In parallel with object-oriented interface the user candirectly insert physical or control equations These last arewritten exactly as in textbooks Starting from individualsubsystems description Modelica-based tool automaticallyperforms many operations first the identification of a set ofdifferential algebraic equations (DAEs) representing the sys-tem under study and then after some additional operationsto simplify the set of equations [15] the conversion in a Cor C++ language compilation to make the final simulationexecutable This task of automatic creation of simulationexecutable requires just a few seconds So changing thesystem to create a new executable (for instance to have moretrains or a different number of ESSs) will require just arepetition of this automatic process Once the executable hasbeen created it can run several times while changing someparameters from a run to another [15]
Last simulator is realised in Matlab-Simscape [14] andentirely developed starting from the tool realised inModelicain this regard the model architecture and equations of the
4 Journal of Advanced Transportation
Train 3
traction circuitequivalent resistors
+ +
Train 2Train 1
ESS1 ESS2
EKEK
RK RK
R1(t) P1u R2(t) P2uR3(t) R4(t)
P1l P2l
Figure 3 Graphical representation of the electrical feeding system
submodels are the same already implemented in ModelicaThus lumped components have been simply rebuilt usingMatlab-Simscape libraries always according to Figure 3 andconsidering for their control part the well-known Matlab-Simulink control libraries
23 Comparisons The three tools were tested by analysingtheir flexibility simulation efficiency and man-machineinterface
Flexibility This aspect was identified in terms of capabilityto model new case studies The Modelica-based tool is ableto rapidly modify the existing models through its object-oriented interface simply by changing or connecting newelements Indeed it is possible to increase or reduce thenumber of the trains running the number and positioningof the electrical feeding substations the eventual storagesystems located along the track the auxiliary adsorption etcIndeed expandability of the model is easily guaranteed alsoby the fact that theModelica-based tool is a unique simulatorunlike those realised in FORTRAN Additionally creation ofthe simulation executable requires just few seconds Thuschanging the system simply requires the rapid repletion ofthis process These characteristics may be retrieved also inMatlab-Simscape although the graphical interface is lessclear and intuitive as only partially object-oriented due tothe need for additional Matlab-Simulink control blocks Onthe other hand the FORTRAN language based model iscompletely different requiring to be fully recompiled at everysmall change [18]
Simulation Efficiency This was evaluated by speed and mem-ory requirements It must be said that the FORTRAN numer-ical solver may be optimised for the considered case studyInstead Modelica and Matlab-Simscape utilise standardnumerical solvers not specifically optimised on the singlecase study The comparison was executed having one systemcharacterised by a single electrical feeding substation and twotrams on the track As already described in [15] at equalnumber of samples and time length simulation theModelica-based tool engaged 115 s requiring 6 MB of memory Thisis nearly the same for Matlab-Simscape which employsabout 20 s to perform a simulation of 1000 s with memoryoccupation of 11 MB In order to consider a proper time forcomplete simulation for FORTRAN calculation code it isneeded to consider the time duration to move between thetwo tools In the case of the considered simulation (ie 1000s) it engaged about 98 s and 20MBofmemory requirements
0100020003000400050006000
0 200 400 600 800 1000
Dist
ance
(m)
time (s)
Figure 4 Pattern profile
Man-Machine Interface The tool developed in Modelicaallows the possibility of making changes in different waysFirst equations can be easily changed by text interface Onthe other hand model structure and related parameters maybe changed by graphical interface [13] This is nearly thesame as utilisation of the Matlab-Simscape tool Regardingthe FORTRAN based model Visual-Basic language is usedto manage input and output data in Train-sim software thesupport of the user friendly macros implemented permitsmodifying parameters in the software although much moreslowly than the others
In conclusion the Modelica-based and Matlab-Simscapetools require high memory requirements but also guaranteemuch more flexibility The FORTRAN based tool developedmany years ago is very useful because it is able to producebenchmark results to be used asmain reference On the otherhand the other two tools allow fast creation of models whosesimulation results need to be carefully verified
3 Case Study
31 Validation The tools have been tested having as referencecase study an existing tramway in Rome The path length isabout 57 km as noticeable from Figure 4 having one singleelectrical feeding substation two trains are on rails one foreach track The main input model parameters are listed inTable 1
The simulation results were evaluated in terms of energyand power flows of the considered tramway In the first exam-ined condition the trams make use of on-board resistors todissipate all the braking energy In the second scenario theysend braking energy on the catenary until the voltage doesnot reach themaximumadmitted value fixed at 800V Finallytrams send braking energy as before but with one storagesystem installed about halfway along the tramway
Journal of Advanced Transportation 5
Table 1 Input model parameters of the system under study
ESS parametersNo load voltage (V) 1680R0(Ω) 013
Number of ESSs 1ESS position (km) 0
Line parametersMax line voltage (V) 1800Nominal line voltage (V) 1650Min line voltage (V) 1100Number of line trunks 6
Line Resistance (Ωkm) 010 010 007007 007 007
Tram parametersFull mass (t) 92Auxiliary power adsorption (kW) 40ED max power (kW) 1242ED max traction force (kN) 96
Track parametersNumber of trams 2Number of stops 14Average distance between stops (km) 04Max speed (ms) 14Track length (km) 54
First simulations were performed without consideringbraking energy recovery As said model parameters wereslightly updated in order tomatch the results provided by thenew tools with the oldest one Figure 5 shows the plot resultsof one train driving on a portion of the considered route Inthis case the tram is not able to send its braking energy alongthe catenary As observable plot results are equivalent amongthe three tools respectively developed in Modelica Matlab-Simscape and FORTRAN
Then simulations have considered the possibility ofrecovering the braking energy In this case parameters actingon control voltage at pantograph have been tuned to performmodulation of the inlet power without overcoming voltagelimits Figure 6 shows plot results of one train under a portionof the considered profile
As noted different blending strategies according to Fig-ure 2 have been implemented Naturally these differenceshave an impact on the total energy absorbed from thenetwork The blending strategy having Vrsquo=09divide095119881max (seeFigure 2) implemented in FORTRAN results in a reductionof the energy recovered through the pantograph respectivelyof 13 and 2 with respect to the dynamical strategyused in Modelica and Matlab-Simscape that allows the fullpowertrain power to be recovered up to 119881max (ie V rsquo=119881max)Then in terms of total energy adsorption from the ESSincrement is of 3 and 06 respectively
Naturally braking energy recovery can be enhancedthrough energy storage systems installed along the route
Table 2 Lithium battery main characteristics
Nominal energy (MWh) 033Nominal capacity (Ah) 200Nominal voltage (V) 1650Number of cells in series 446Max allowed current (A) 2000Charging-discharging efficiency 09
Table 3 ESS energy consumption
ESS working day daily energy (MWh) 158ESS holiday daily energy (kWh) 111ESS Annual energy (MWh) 5386
In this way one storage system has been introduced withthe main aim of validating the tools under test also in thenew considered system configuration The lithium battery ispositioned about halfway along the tramway (ie about 38km from the terminal) whose characteristics are in Table 2
Relation among the nominal energy and nominal capac-ity is given by
119864119899= 119899119904119881119899119888119862119899 (5)
where 119899119904is the number of cells in series 119864
119899is the nominal
energy 119881119899119888is the nominal cell voltage and 119862
119899is the nominal
capacity Number of cells and nominal capacity were selectedin order to exactly reproduce through the newest tools thebattery SOC (State-of-Charge) and the power profile withrespect to the FORTRAN based one Results are shown inFigure 7
Installation of one storage system can significantly reducethe energy delivered by the electrical feeding substations(ESSs) The considered simplified case study with two trainsobviously cannot correctly evaluate the cost-effectivenessof the considered solution Thus the full analysis will bedescribed in the next section following the same approachalready considered by the authors in [12 13]
32 Energy Saving Evaluation Energy saving evaluationhas been performed by considering an experimental mea-surement campaign carried on by the authors Results aresummarized in Table 3 where the daily energy delivered fromthe electrical substation (ESS) has been measured duringone typical school working day and one typical holidayday Starting from that evaluation of the annual electricalenergy demand has been performed simply by consideringthe number of working days and holidays over the year givenby the tramline operator
Therefore the Modelica-based tool presented before hasbeen used to exactly reproduce the energy consumptionshown in Table 3 To do that different numbers of operatingtrams along the day have been considered from timetablerespectively 24 during peak hours typically concentrated inthe middle part of the day and 12 in low load hours mainlyin the early morning and in the late evening Real numberof trams in operation may be slightly different However the
6 Journal of Advanced Transportation
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
Tram
1 V
olta
ge (V
)
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
times106
Figure 5 Plot results trams without braking energy recovery
daily distribution has been chosen to exactly reproduce themeasured consumption
After simulating exactly the same level of the measuredenergy demand ie obtaining the daily consumption shownin Table 3 simulations have been repeated by including thestorage system described in Table 2 The delivered energy bythe ESS was then compared with the previous one calculatedin absence of the storage system The reduction in annualenergy consumption was about 152 moving from 5386MWh up to 4570MWh
The cost-effectiveness of the proposed solution has beeninvestigated by considering the initial cash outlay due theintroduction of the storage system with respect to theannual return of the investment due to the above-mentionedelectrical energy saving The initial cash outlay due to thestorage system and its balance of plant has been calculatedby considering a value of 500 eurokWh including cells BMSand battery packaging As discussed in detail in [19] thepresence of a DCDC converter may be required For thisa fixed cost of 60 keuro in analogue way to the experiencepresented in [19] has been used Finally the current industrialuser price of energy in Italy was evaluated considering anaverage value of 150 euroMWh In terms of maintenancecosts for the reasons already explained in [19] the storagehas been considered able to cover the whole life of theplant
Table 4 Economic benefit analysis
Storage system cost (keuro) 2261Annual energy saving (keuro) 1225NPV (keuro) 7382PBT (y) 2
Main objective of the analysis was related to the evalua-tion of the net present value (NPV) and of the payback time(PBT) for a whole life of ten years and an interest rate of 4Results are shown in Table 4
The results show that installation of a stationary storagesystem may guarantee a payback time within just two yearThese numbers are so favourable not to be affected by anypossible storage substitution during the plant life Indeed thecost-effectiveness of the proposed solution has been clearlydemonstrated
4 Conclusions
This paper has demonstrated how innovative languages andsoftware allow rapid creation of numerical models havingelectrical mechanical and control parts to be simulated
The Modelica-based model was realised through thecommercial Dymola tool However it could run inside any
Journal of Advanced Transportation 7
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
2000
Tram
1 V
olta
ge (V
)
Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
times106
Figure 6 Plot results trams with braking energy recovery
0 100 200 300 400 500 600 700 800 900 1000Time (s)
04
045
05
055
RESS
SO
C (p
u)
Simscape Dymola Fortran
Simscape Dymola Fortran
0 100 200 300 400 500 600 700 800 900 1000Time (s)
minus1
0
1
2
RESS
pow
er (W
)
times106
Figure 7 Plot results system equipped with storage
8 Journal of Advanced Transportation
other Modelica-compliant tool They proved to be fast withrelatively small memory occupation Nearly the same can besaid about Matlab-Simscape although characterised by lessflexibility due to the fact that it is not inspired by an open-source language platform The quality has been verified bycomparing results with those obtained using a well validatedFORTRAN based simulator
With reference to an existing case study the cost-effectiveness due to the utilisation of stationary storagesystem to enhance braking energy recovery has been clearlydemonstrated since on a high-traffic tramway with highnumber of stops payback time has been reached within onlytwo years Naturally it is of great importance in order to cor-rectly achieve energy saving for the considered application topreliminary calibrate the considered tool on actual electricalenergy consumption experimentally measured
As future direction of this work it is also possible toconsider the extension of the considered methodology toother tramlines in operation in order to investigate thepotential cost-effectiveness of similar or different energysaving solutions
Data Availability
The data used to support the findings of this study areincluded within the article
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] A Gonzalez-Gil R Palacin P Batty and J P Powell ldquoA systemsapproach to reduce urban rail energy consumptionrdquo EnergyConversion and Management vol 80 pp 509ndash524 2014
[2] A Gonzalez-Gil R Palacin and P Batty ldquoSustainable urban railsystems Strategies and technologies for optimalmanagement ofregenerative braking energyrdquo Energy Conversion and Manage-ment vol 75 pp 374ndash388 2013
[3] M C Falvo R Lamedica and A Ruvio ldquoEnergy storage appli-cation in trolley-buses lines for a sustainable urban mobilityrdquoin Proceedings of the 2012 Electrical Systems for Aircraft Railwayand Ship Propulsion ESARS 2012 Italy October 2012
[4] M C Falvo R Lamedica and A Ruvio ldquoAn environmentalsustainable transport system A trolley-buses Line for Cosenzacityrdquo in Proceedings of the 21st International Symposium onPower Electronics Electrical Drives Automation and MotionSPEEDAM 2012 pp 1479ndash1485 Italy June 2012
[5] M Berger C Lavertu I Kocar and J Mahseredjian ldquoProposalof a time-domain platform for short-circuit protection analysisin rapid transit train DC auxiliary systemsrdquo IEEE Transactionson Industry Applications vol 52 no 6 pp 5295ndash5304 2016
[6] A Ibrahem A Elrayyah Y Sozer and J A De Abreu-GarcialdquoDC railway system emulator for stray current and touchvoltage predictionrdquo IEEE Transactions on Industry Applicationsvol 53 no 1 pp 439ndash446 2017
[7] A Capasso R Lamedica andC Penna ldquoEnergy regeneration intransportation systemmethodologies for powernetworks simu-lationrdquo in Proceedings of the 4th IFACIFIPIFORS international
conference on control in transportation system Baden GermanyApril 1983
[8] A Capasso G Giannini R Lamedica and A Ruvio ldquoEco-friendly urban transport systems Comparison between energydemands of the trolleybus and tram systemsrdquo Ingegneria Fer-roviaria vol 69 no 4 pp 329ndash347 2014
[9] A Adinolfi R Lamedica C Modesto A Prudenzi and SVimercati ldquoExperimental assessment of energy saving due totrains regenerative braking in an electrified subway linerdquo IEEETransactions on Power Delivery vol 13 no 4 pp 1536ndash15411998
[10] Dymola ldquoDynamic Modeling Laboratoryrdquo documentationavailable from httpwww3dscomproductsservicescatiaproductsdymola
[11] P Fritzson Introduction toModeling and Simulation of Technicaland Physical Systems with Modelica Wiley-IEEE Press 2011
[12] M Ceraolo R Giglioli G Lutzemberger M Conte and MPasquali ldquoUse of electrochemical storage in substations toenhance energy and cost efficiency of tramwaysrdquo in Proceedingsof the 2013 AEIT Annual Conference Innovation and Scientificand Technical Culture for Development AEIT 2013 MondelloItaly October 2013
[13] S Barsali P Bolognesi M Ceraolo M Funaioli and G Lu-tzembergerCyber-Physical Modelling of Railroad Vehicle SystemusingModelica Simulation Language vol 104 Railways AjaccioCorsica France 2014
[14] ldquoMatlab-Simscaperdquo Official site httpswwwmathworkscomproductssimscapehtml
[15] A Capasso R Lamedica A Ruvio M Ceraolo and GLutzemberger ldquoModelling and simulation of electric urbantransportation systems with energy storagerdquo in Proceedings ofthe 16th International Conference on Environment and ElectricalEngineering (EEEIC rsquo16) Florence Italy June 2016
[16] ldquoCEI EN 50388 Railway applications - Power supply and rollingstockrdquo 2012
[17] M E El-Hawary and O K Wellon ldquoThe alpha-modified quasi-second order Newton-Raphson method for load flow solutionsin rectangular formrdquo IEEE Transactions on Power Apparatusand Systems vol 101 no 4 pp 854ndash866 1982
[18] A Capasso and R Lamedica Simulation of The Operation ofElectric Traction Networks CNR - Progetto finalizzato trasportiBologna Italy 1984
[19] M Ceraolo R Giglioli G Lutzemberger and A BechinildquoCost effective storage for energy saving in feeding systems oftramwaysrdquo in Proceedings of the 2014 IEEE International ElectricVehicle Conference IEVC 2014 IEEE Florence Italy December2014
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Journal of Advanced Transportation 3
V node Cond NodeTraction
Braking
P
V
6GCH 6lowast 6 6GR
Vlowast = 08 Vmax
6GCHlt6lt6lowast
6lowastlt6lt6GR
6lt6GR
6=6GR
I=costle)GR
P=cost
P=costP=0
Figure 2 Voltage-power characteristic during traction and brakingstages
For each electrical substation the software calculatedelectrical parameters about power current etc including theamount of recoverable energy during braking and the voltagealong the line
The procedure for determining the matrix coefficients ofthe admittance constituting the traction line refers to electri-cal networks in permanent sinusoidal regime and thereforeto complex admittance and impedanceThe software dealingwithDCnetworks considers all themagnitudes to be real theadmittance as conductance and the impedance as resistance
Regarding numerical solving methods Newton-Raphsonor Gauss-Seidel method is typically applied Some variantsof Newton-Raphson method particularly suitable for small-and medium-size networks can be also considered in orderto improve conditions of convergence and to reduce numberof iterations [17]
22 Modelica and Matlab-Simscape Based Model As saidthe simulation tool realised in Dymola is based on Mod-elica language [11] It is a single simulator performing thesame functionalities of the two previous ones written inFORTRAN to the advantage of accuracy and flexibility aswill be analysed later As anticipated Modelica language iscyber-physical and allows simulating complex systems iehaving mechanical electrical thermal control subsystemsFollowing the proposed approach the ESSs were simulatedthrough lumped components linked in a graphical wayas visible in Figure 3 As this study shows only the DCcomponent is of interest to evaluate effects of harmonics eachsubstation has been modelled through its well-known DCequivalent Naturally the system is time-variant In fact thecontact-line configuration varies with time since the trainposition varies In fact resistor values change depending onthe train position according to the following expressions
1198771= (1 minus 120575) 119877
119905119900119905
1198772= 120575119877119905119900119905
120575 =(1198752minus 119909)
11987112
(3)
whereR1is the left side line resistance R
2is the right side line
resistance and 120575 is the ratio of the distance between the trainposition (x) and ESS2 position (P
2) and the distance between
ESS1 and ESS2 substations (L12)
In addition the electrical feeding substations in operationare subjected to variation when a train moves from asection to another along the track This difficulty can beeasily addressed in Modelica considering the possibility ofchanging the system equations after some events happen[12 13]
Finally modelling of trains requires modelling of theelectric drive resistance forces and the driverrsquos behaviourElectric drive ismodelled as a system able to produce the trac-tive force as required by the driver within the allowed forceand power limits generating some power losses expressed asa function of the mechanical speed and the required forceThen each train must avoid feeding power to the catenarywhen this would cause the line voltage to become too largeand a controller of the DC power must be implementedA much more sophisticated control strategy has been usedhaving feedback on the instantaneous pantograph voltage andmodulating the braking power conveyed along the catenaryin order to avoid reaching instantaneously the upper allowedlimit Resistance to movement has been modelled using theformula including aerodynamic drag and rolling resistance
119877 = 119898119892 (119891119903cos 120572 + sin 120572) + 119860119881 + 1198611198812 (4)
whereR is the global resistance force acting against the vehiclemovement 120572 is the angle between the track and the horizon-tal plane m is the vehicle mass 119891
119903is the rolling resistance
coefficient A and B are empirical positive numbers takinginto account the lateral and front aerodynamic resistance andV is the train speed About the driver it can be representedsimply by a proportional controller in which the referencetractive force is proportional to the error between the actualand the reference speed
Further details regarding different submodel and controllogic in particular the blending strategy depicted in Figure 2are widely described also in [15]
In parallel with object-oriented interface the user candirectly insert physical or control equations These last arewritten exactly as in textbooks Starting from individualsubsystems description Modelica-based tool automaticallyperforms many operations first the identification of a set ofdifferential algebraic equations (DAEs) representing the sys-tem under study and then after some additional operationsto simplify the set of equations [15] the conversion in a Cor C++ language compilation to make the final simulationexecutable This task of automatic creation of simulationexecutable requires just a few seconds So changing thesystem to create a new executable (for instance to have moretrains or a different number of ESSs) will require just arepetition of this automatic process Once the executable hasbeen created it can run several times while changing someparameters from a run to another [15]
Last simulator is realised in Matlab-Simscape [14] andentirely developed starting from the tool realised inModelicain this regard the model architecture and equations of the
4 Journal of Advanced Transportation
Train 3
traction circuitequivalent resistors
+ +
Train 2Train 1
ESS1 ESS2
EKEK
RK RK
R1(t) P1u R2(t) P2uR3(t) R4(t)
P1l P2l
Figure 3 Graphical representation of the electrical feeding system
submodels are the same already implemented in ModelicaThus lumped components have been simply rebuilt usingMatlab-Simscape libraries always according to Figure 3 andconsidering for their control part the well-known Matlab-Simulink control libraries
23 Comparisons The three tools were tested by analysingtheir flexibility simulation efficiency and man-machineinterface
Flexibility This aspect was identified in terms of capabilityto model new case studies The Modelica-based tool is ableto rapidly modify the existing models through its object-oriented interface simply by changing or connecting newelements Indeed it is possible to increase or reduce thenumber of the trains running the number and positioningof the electrical feeding substations the eventual storagesystems located along the track the auxiliary adsorption etcIndeed expandability of the model is easily guaranteed alsoby the fact that theModelica-based tool is a unique simulatorunlike those realised in FORTRAN Additionally creation ofthe simulation executable requires just few seconds Thuschanging the system simply requires the rapid repletion ofthis process These characteristics may be retrieved also inMatlab-Simscape although the graphical interface is lessclear and intuitive as only partially object-oriented due tothe need for additional Matlab-Simulink control blocks Onthe other hand the FORTRAN language based model iscompletely different requiring to be fully recompiled at everysmall change [18]
Simulation Efficiency This was evaluated by speed and mem-ory requirements It must be said that the FORTRAN numer-ical solver may be optimised for the considered case studyInstead Modelica and Matlab-Simscape utilise standardnumerical solvers not specifically optimised on the singlecase study The comparison was executed having one systemcharacterised by a single electrical feeding substation and twotrams on the track As already described in [15] at equalnumber of samples and time length simulation theModelica-based tool engaged 115 s requiring 6 MB of memory Thisis nearly the same for Matlab-Simscape which employsabout 20 s to perform a simulation of 1000 s with memoryoccupation of 11 MB In order to consider a proper time forcomplete simulation for FORTRAN calculation code it isneeded to consider the time duration to move between thetwo tools In the case of the considered simulation (ie 1000s) it engaged about 98 s and 20MBofmemory requirements
0100020003000400050006000
0 200 400 600 800 1000
Dist
ance
(m)
time (s)
Figure 4 Pattern profile
Man-Machine Interface The tool developed in Modelicaallows the possibility of making changes in different waysFirst equations can be easily changed by text interface Onthe other hand model structure and related parameters maybe changed by graphical interface [13] This is nearly thesame as utilisation of the Matlab-Simscape tool Regardingthe FORTRAN based model Visual-Basic language is usedto manage input and output data in Train-sim software thesupport of the user friendly macros implemented permitsmodifying parameters in the software although much moreslowly than the others
In conclusion the Modelica-based and Matlab-Simscapetools require high memory requirements but also guaranteemuch more flexibility The FORTRAN based tool developedmany years ago is very useful because it is able to producebenchmark results to be used asmain reference On the otherhand the other two tools allow fast creation of models whosesimulation results need to be carefully verified
3 Case Study
31 Validation The tools have been tested having as referencecase study an existing tramway in Rome The path length isabout 57 km as noticeable from Figure 4 having one singleelectrical feeding substation two trains are on rails one foreach track The main input model parameters are listed inTable 1
The simulation results were evaluated in terms of energyand power flows of the considered tramway In the first exam-ined condition the trams make use of on-board resistors todissipate all the braking energy In the second scenario theysend braking energy on the catenary until the voltage doesnot reach themaximumadmitted value fixed at 800V Finallytrams send braking energy as before but with one storagesystem installed about halfway along the tramway
Journal of Advanced Transportation 5
Table 1 Input model parameters of the system under study
ESS parametersNo load voltage (V) 1680R0(Ω) 013
Number of ESSs 1ESS position (km) 0
Line parametersMax line voltage (V) 1800Nominal line voltage (V) 1650Min line voltage (V) 1100Number of line trunks 6
Line Resistance (Ωkm) 010 010 007007 007 007
Tram parametersFull mass (t) 92Auxiliary power adsorption (kW) 40ED max power (kW) 1242ED max traction force (kN) 96
Track parametersNumber of trams 2Number of stops 14Average distance between stops (km) 04Max speed (ms) 14Track length (km) 54
First simulations were performed without consideringbraking energy recovery As said model parameters wereslightly updated in order tomatch the results provided by thenew tools with the oldest one Figure 5 shows the plot resultsof one train driving on a portion of the considered route Inthis case the tram is not able to send its braking energy alongthe catenary As observable plot results are equivalent amongthe three tools respectively developed in Modelica Matlab-Simscape and FORTRAN
Then simulations have considered the possibility ofrecovering the braking energy In this case parameters actingon control voltage at pantograph have been tuned to performmodulation of the inlet power without overcoming voltagelimits Figure 6 shows plot results of one train under a portionof the considered profile
As noted different blending strategies according to Fig-ure 2 have been implemented Naturally these differenceshave an impact on the total energy absorbed from thenetwork The blending strategy having Vrsquo=09divide095119881max (seeFigure 2) implemented in FORTRAN results in a reductionof the energy recovered through the pantograph respectivelyof 13 and 2 with respect to the dynamical strategyused in Modelica and Matlab-Simscape that allows the fullpowertrain power to be recovered up to 119881max (ie V rsquo=119881max)Then in terms of total energy adsorption from the ESSincrement is of 3 and 06 respectively
Naturally braking energy recovery can be enhancedthrough energy storage systems installed along the route
Table 2 Lithium battery main characteristics
Nominal energy (MWh) 033Nominal capacity (Ah) 200Nominal voltage (V) 1650Number of cells in series 446Max allowed current (A) 2000Charging-discharging efficiency 09
Table 3 ESS energy consumption
ESS working day daily energy (MWh) 158ESS holiday daily energy (kWh) 111ESS Annual energy (MWh) 5386
In this way one storage system has been introduced withthe main aim of validating the tools under test also in thenew considered system configuration The lithium battery ispositioned about halfway along the tramway (ie about 38km from the terminal) whose characteristics are in Table 2
Relation among the nominal energy and nominal capac-ity is given by
119864119899= 119899119904119881119899119888119862119899 (5)
where 119899119904is the number of cells in series 119864
119899is the nominal
energy 119881119899119888is the nominal cell voltage and 119862
119899is the nominal
capacity Number of cells and nominal capacity were selectedin order to exactly reproduce through the newest tools thebattery SOC (State-of-Charge) and the power profile withrespect to the FORTRAN based one Results are shown inFigure 7
Installation of one storage system can significantly reducethe energy delivered by the electrical feeding substations(ESSs) The considered simplified case study with two trainsobviously cannot correctly evaluate the cost-effectivenessof the considered solution Thus the full analysis will bedescribed in the next section following the same approachalready considered by the authors in [12 13]
32 Energy Saving Evaluation Energy saving evaluationhas been performed by considering an experimental mea-surement campaign carried on by the authors Results aresummarized in Table 3 where the daily energy delivered fromthe electrical substation (ESS) has been measured duringone typical school working day and one typical holidayday Starting from that evaluation of the annual electricalenergy demand has been performed simply by consideringthe number of working days and holidays over the year givenby the tramline operator
Therefore the Modelica-based tool presented before hasbeen used to exactly reproduce the energy consumptionshown in Table 3 To do that different numbers of operatingtrams along the day have been considered from timetablerespectively 24 during peak hours typically concentrated inthe middle part of the day and 12 in low load hours mainlyin the early morning and in the late evening Real numberof trams in operation may be slightly different However the
6 Journal of Advanced Transportation
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
Tram
1 V
olta
ge (V
)
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
times106
Figure 5 Plot results trams without braking energy recovery
daily distribution has been chosen to exactly reproduce themeasured consumption
After simulating exactly the same level of the measuredenergy demand ie obtaining the daily consumption shownin Table 3 simulations have been repeated by including thestorage system described in Table 2 The delivered energy bythe ESS was then compared with the previous one calculatedin absence of the storage system The reduction in annualenergy consumption was about 152 moving from 5386MWh up to 4570MWh
The cost-effectiveness of the proposed solution has beeninvestigated by considering the initial cash outlay due theintroduction of the storage system with respect to theannual return of the investment due to the above-mentionedelectrical energy saving The initial cash outlay due to thestorage system and its balance of plant has been calculatedby considering a value of 500 eurokWh including cells BMSand battery packaging As discussed in detail in [19] thepresence of a DCDC converter may be required For thisa fixed cost of 60 keuro in analogue way to the experiencepresented in [19] has been used Finally the current industrialuser price of energy in Italy was evaluated considering anaverage value of 150 euroMWh In terms of maintenancecosts for the reasons already explained in [19] the storagehas been considered able to cover the whole life of theplant
Table 4 Economic benefit analysis
Storage system cost (keuro) 2261Annual energy saving (keuro) 1225NPV (keuro) 7382PBT (y) 2
Main objective of the analysis was related to the evalua-tion of the net present value (NPV) and of the payback time(PBT) for a whole life of ten years and an interest rate of 4Results are shown in Table 4
The results show that installation of a stationary storagesystem may guarantee a payback time within just two yearThese numbers are so favourable not to be affected by anypossible storage substitution during the plant life Indeed thecost-effectiveness of the proposed solution has been clearlydemonstrated
4 Conclusions
This paper has demonstrated how innovative languages andsoftware allow rapid creation of numerical models havingelectrical mechanical and control parts to be simulated
The Modelica-based model was realised through thecommercial Dymola tool However it could run inside any
Journal of Advanced Transportation 7
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
2000
Tram
1 V
olta
ge (V
)
Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
times106
Figure 6 Plot results trams with braking energy recovery
0 100 200 300 400 500 600 700 800 900 1000Time (s)
04
045
05
055
RESS
SO
C (p
u)
Simscape Dymola Fortran
Simscape Dymola Fortran
0 100 200 300 400 500 600 700 800 900 1000Time (s)
minus1
0
1
2
RESS
pow
er (W
)
times106
Figure 7 Plot results system equipped with storage
8 Journal of Advanced Transportation
other Modelica-compliant tool They proved to be fast withrelatively small memory occupation Nearly the same can besaid about Matlab-Simscape although characterised by lessflexibility due to the fact that it is not inspired by an open-source language platform The quality has been verified bycomparing results with those obtained using a well validatedFORTRAN based simulator
With reference to an existing case study the cost-effectiveness due to the utilisation of stationary storagesystem to enhance braking energy recovery has been clearlydemonstrated since on a high-traffic tramway with highnumber of stops payback time has been reached within onlytwo years Naturally it is of great importance in order to cor-rectly achieve energy saving for the considered application topreliminary calibrate the considered tool on actual electricalenergy consumption experimentally measured
As future direction of this work it is also possible toconsider the extension of the considered methodology toother tramlines in operation in order to investigate thepotential cost-effectiveness of similar or different energysaving solutions
Data Availability
The data used to support the findings of this study areincluded within the article
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] A Gonzalez-Gil R Palacin P Batty and J P Powell ldquoA systemsapproach to reduce urban rail energy consumptionrdquo EnergyConversion and Management vol 80 pp 509ndash524 2014
[2] A Gonzalez-Gil R Palacin and P Batty ldquoSustainable urban railsystems Strategies and technologies for optimalmanagement ofregenerative braking energyrdquo Energy Conversion and Manage-ment vol 75 pp 374ndash388 2013
[3] M C Falvo R Lamedica and A Ruvio ldquoEnergy storage appli-cation in trolley-buses lines for a sustainable urban mobilityrdquoin Proceedings of the 2012 Electrical Systems for Aircraft Railwayand Ship Propulsion ESARS 2012 Italy October 2012
[4] M C Falvo R Lamedica and A Ruvio ldquoAn environmentalsustainable transport system A trolley-buses Line for Cosenzacityrdquo in Proceedings of the 21st International Symposium onPower Electronics Electrical Drives Automation and MotionSPEEDAM 2012 pp 1479ndash1485 Italy June 2012
[5] M Berger C Lavertu I Kocar and J Mahseredjian ldquoProposalof a time-domain platform for short-circuit protection analysisin rapid transit train DC auxiliary systemsrdquo IEEE Transactionson Industry Applications vol 52 no 6 pp 5295ndash5304 2016
[6] A Ibrahem A Elrayyah Y Sozer and J A De Abreu-GarcialdquoDC railway system emulator for stray current and touchvoltage predictionrdquo IEEE Transactions on Industry Applicationsvol 53 no 1 pp 439ndash446 2017
[7] A Capasso R Lamedica andC Penna ldquoEnergy regeneration intransportation systemmethodologies for powernetworks simu-lationrdquo in Proceedings of the 4th IFACIFIPIFORS international
conference on control in transportation system Baden GermanyApril 1983
[8] A Capasso G Giannini R Lamedica and A Ruvio ldquoEco-friendly urban transport systems Comparison between energydemands of the trolleybus and tram systemsrdquo Ingegneria Fer-roviaria vol 69 no 4 pp 329ndash347 2014
[9] A Adinolfi R Lamedica C Modesto A Prudenzi and SVimercati ldquoExperimental assessment of energy saving due totrains regenerative braking in an electrified subway linerdquo IEEETransactions on Power Delivery vol 13 no 4 pp 1536ndash15411998
[10] Dymola ldquoDynamic Modeling Laboratoryrdquo documentationavailable from httpwww3dscomproductsservicescatiaproductsdymola
[11] P Fritzson Introduction toModeling and Simulation of Technicaland Physical Systems with Modelica Wiley-IEEE Press 2011
[12] M Ceraolo R Giglioli G Lutzemberger M Conte and MPasquali ldquoUse of electrochemical storage in substations toenhance energy and cost efficiency of tramwaysrdquo in Proceedingsof the 2013 AEIT Annual Conference Innovation and Scientificand Technical Culture for Development AEIT 2013 MondelloItaly October 2013
[13] S Barsali P Bolognesi M Ceraolo M Funaioli and G Lu-tzembergerCyber-Physical Modelling of Railroad Vehicle SystemusingModelica Simulation Language vol 104 Railways AjaccioCorsica France 2014
[14] ldquoMatlab-Simscaperdquo Official site httpswwwmathworkscomproductssimscapehtml
[15] A Capasso R Lamedica A Ruvio M Ceraolo and GLutzemberger ldquoModelling and simulation of electric urbantransportation systems with energy storagerdquo in Proceedings ofthe 16th International Conference on Environment and ElectricalEngineering (EEEIC rsquo16) Florence Italy June 2016
[16] ldquoCEI EN 50388 Railway applications - Power supply and rollingstockrdquo 2012
[17] M E El-Hawary and O K Wellon ldquoThe alpha-modified quasi-second order Newton-Raphson method for load flow solutionsin rectangular formrdquo IEEE Transactions on Power Apparatusand Systems vol 101 no 4 pp 854ndash866 1982
[18] A Capasso and R Lamedica Simulation of The Operation ofElectric Traction Networks CNR - Progetto finalizzato trasportiBologna Italy 1984
[19] M Ceraolo R Giglioli G Lutzemberger and A BechinildquoCost effective storage for energy saving in feeding systems oftramwaysrdquo in Proceedings of the 2014 IEEE International ElectricVehicle Conference IEVC 2014 IEEE Florence Italy December2014
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
4 Journal of Advanced Transportation
Train 3
traction circuitequivalent resistors
+ +
Train 2Train 1
ESS1 ESS2
EKEK
RK RK
R1(t) P1u R2(t) P2uR3(t) R4(t)
P1l P2l
Figure 3 Graphical representation of the electrical feeding system
submodels are the same already implemented in ModelicaThus lumped components have been simply rebuilt usingMatlab-Simscape libraries always according to Figure 3 andconsidering for their control part the well-known Matlab-Simulink control libraries
23 Comparisons The three tools were tested by analysingtheir flexibility simulation efficiency and man-machineinterface
Flexibility This aspect was identified in terms of capabilityto model new case studies The Modelica-based tool is ableto rapidly modify the existing models through its object-oriented interface simply by changing or connecting newelements Indeed it is possible to increase or reduce thenumber of the trains running the number and positioningof the electrical feeding substations the eventual storagesystems located along the track the auxiliary adsorption etcIndeed expandability of the model is easily guaranteed alsoby the fact that theModelica-based tool is a unique simulatorunlike those realised in FORTRAN Additionally creation ofthe simulation executable requires just few seconds Thuschanging the system simply requires the rapid repletion ofthis process These characteristics may be retrieved also inMatlab-Simscape although the graphical interface is lessclear and intuitive as only partially object-oriented due tothe need for additional Matlab-Simulink control blocks Onthe other hand the FORTRAN language based model iscompletely different requiring to be fully recompiled at everysmall change [18]
Simulation Efficiency This was evaluated by speed and mem-ory requirements It must be said that the FORTRAN numer-ical solver may be optimised for the considered case studyInstead Modelica and Matlab-Simscape utilise standardnumerical solvers not specifically optimised on the singlecase study The comparison was executed having one systemcharacterised by a single electrical feeding substation and twotrams on the track As already described in [15] at equalnumber of samples and time length simulation theModelica-based tool engaged 115 s requiring 6 MB of memory Thisis nearly the same for Matlab-Simscape which employsabout 20 s to perform a simulation of 1000 s with memoryoccupation of 11 MB In order to consider a proper time forcomplete simulation for FORTRAN calculation code it isneeded to consider the time duration to move between thetwo tools In the case of the considered simulation (ie 1000s) it engaged about 98 s and 20MBofmemory requirements
0100020003000400050006000
0 200 400 600 800 1000
Dist
ance
(m)
time (s)
Figure 4 Pattern profile
Man-Machine Interface The tool developed in Modelicaallows the possibility of making changes in different waysFirst equations can be easily changed by text interface Onthe other hand model structure and related parameters maybe changed by graphical interface [13] This is nearly thesame as utilisation of the Matlab-Simscape tool Regardingthe FORTRAN based model Visual-Basic language is usedto manage input and output data in Train-sim software thesupport of the user friendly macros implemented permitsmodifying parameters in the software although much moreslowly than the others
In conclusion the Modelica-based and Matlab-Simscapetools require high memory requirements but also guaranteemuch more flexibility The FORTRAN based tool developedmany years ago is very useful because it is able to producebenchmark results to be used asmain reference On the otherhand the other two tools allow fast creation of models whosesimulation results need to be carefully verified
3 Case Study
31 Validation The tools have been tested having as referencecase study an existing tramway in Rome The path length isabout 57 km as noticeable from Figure 4 having one singleelectrical feeding substation two trains are on rails one foreach track The main input model parameters are listed inTable 1
The simulation results were evaluated in terms of energyand power flows of the considered tramway In the first exam-ined condition the trams make use of on-board resistors todissipate all the braking energy In the second scenario theysend braking energy on the catenary until the voltage doesnot reach themaximumadmitted value fixed at 800V Finallytrams send braking energy as before but with one storagesystem installed about halfway along the tramway
Journal of Advanced Transportation 5
Table 1 Input model parameters of the system under study
ESS parametersNo load voltage (V) 1680R0(Ω) 013
Number of ESSs 1ESS position (km) 0
Line parametersMax line voltage (V) 1800Nominal line voltage (V) 1650Min line voltage (V) 1100Number of line trunks 6
Line Resistance (Ωkm) 010 010 007007 007 007
Tram parametersFull mass (t) 92Auxiliary power adsorption (kW) 40ED max power (kW) 1242ED max traction force (kN) 96
Track parametersNumber of trams 2Number of stops 14Average distance between stops (km) 04Max speed (ms) 14Track length (km) 54
First simulations were performed without consideringbraking energy recovery As said model parameters wereslightly updated in order tomatch the results provided by thenew tools with the oldest one Figure 5 shows the plot resultsof one train driving on a portion of the considered route Inthis case the tram is not able to send its braking energy alongthe catenary As observable plot results are equivalent amongthe three tools respectively developed in Modelica Matlab-Simscape and FORTRAN
Then simulations have considered the possibility ofrecovering the braking energy In this case parameters actingon control voltage at pantograph have been tuned to performmodulation of the inlet power without overcoming voltagelimits Figure 6 shows plot results of one train under a portionof the considered profile
As noted different blending strategies according to Fig-ure 2 have been implemented Naturally these differenceshave an impact on the total energy absorbed from thenetwork The blending strategy having Vrsquo=09divide095119881max (seeFigure 2) implemented in FORTRAN results in a reductionof the energy recovered through the pantograph respectivelyof 13 and 2 with respect to the dynamical strategyused in Modelica and Matlab-Simscape that allows the fullpowertrain power to be recovered up to 119881max (ie V rsquo=119881max)Then in terms of total energy adsorption from the ESSincrement is of 3 and 06 respectively
Naturally braking energy recovery can be enhancedthrough energy storage systems installed along the route
Table 2 Lithium battery main characteristics
Nominal energy (MWh) 033Nominal capacity (Ah) 200Nominal voltage (V) 1650Number of cells in series 446Max allowed current (A) 2000Charging-discharging efficiency 09
Table 3 ESS energy consumption
ESS working day daily energy (MWh) 158ESS holiday daily energy (kWh) 111ESS Annual energy (MWh) 5386
In this way one storage system has been introduced withthe main aim of validating the tools under test also in thenew considered system configuration The lithium battery ispositioned about halfway along the tramway (ie about 38km from the terminal) whose characteristics are in Table 2
Relation among the nominal energy and nominal capac-ity is given by
119864119899= 119899119904119881119899119888119862119899 (5)
where 119899119904is the number of cells in series 119864
119899is the nominal
energy 119881119899119888is the nominal cell voltage and 119862
119899is the nominal
capacity Number of cells and nominal capacity were selectedin order to exactly reproduce through the newest tools thebattery SOC (State-of-Charge) and the power profile withrespect to the FORTRAN based one Results are shown inFigure 7
Installation of one storage system can significantly reducethe energy delivered by the electrical feeding substations(ESSs) The considered simplified case study with two trainsobviously cannot correctly evaluate the cost-effectivenessof the considered solution Thus the full analysis will bedescribed in the next section following the same approachalready considered by the authors in [12 13]
32 Energy Saving Evaluation Energy saving evaluationhas been performed by considering an experimental mea-surement campaign carried on by the authors Results aresummarized in Table 3 where the daily energy delivered fromthe electrical substation (ESS) has been measured duringone typical school working day and one typical holidayday Starting from that evaluation of the annual electricalenergy demand has been performed simply by consideringthe number of working days and holidays over the year givenby the tramline operator
Therefore the Modelica-based tool presented before hasbeen used to exactly reproduce the energy consumptionshown in Table 3 To do that different numbers of operatingtrams along the day have been considered from timetablerespectively 24 during peak hours typically concentrated inthe middle part of the day and 12 in low load hours mainlyin the early morning and in the late evening Real numberof trams in operation may be slightly different However the
6 Journal of Advanced Transportation
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
Tram
1 V
olta
ge (V
)
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
times106
Figure 5 Plot results trams without braking energy recovery
daily distribution has been chosen to exactly reproduce themeasured consumption
After simulating exactly the same level of the measuredenergy demand ie obtaining the daily consumption shownin Table 3 simulations have been repeated by including thestorage system described in Table 2 The delivered energy bythe ESS was then compared with the previous one calculatedin absence of the storage system The reduction in annualenergy consumption was about 152 moving from 5386MWh up to 4570MWh
The cost-effectiveness of the proposed solution has beeninvestigated by considering the initial cash outlay due theintroduction of the storage system with respect to theannual return of the investment due to the above-mentionedelectrical energy saving The initial cash outlay due to thestorage system and its balance of plant has been calculatedby considering a value of 500 eurokWh including cells BMSand battery packaging As discussed in detail in [19] thepresence of a DCDC converter may be required For thisa fixed cost of 60 keuro in analogue way to the experiencepresented in [19] has been used Finally the current industrialuser price of energy in Italy was evaluated considering anaverage value of 150 euroMWh In terms of maintenancecosts for the reasons already explained in [19] the storagehas been considered able to cover the whole life of theplant
Table 4 Economic benefit analysis
Storage system cost (keuro) 2261Annual energy saving (keuro) 1225NPV (keuro) 7382PBT (y) 2
Main objective of the analysis was related to the evalua-tion of the net present value (NPV) and of the payback time(PBT) for a whole life of ten years and an interest rate of 4Results are shown in Table 4
The results show that installation of a stationary storagesystem may guarantee a payback time within just two yearThese numbers are so favourable not to be affected by anypossible storage substitution during the plant life Indeed thecost-effectiveness of the proposed solution has been clearlydemonstrated
4 Conclusions
This paper has demonstrated how innovative languages andsoftware allow rapid creation of numerical models havingelectrical mechanical and control parts to be simulated
The Modelica-based model was realised through thecommercial Dymola tool However it could run inside any
Journal of Advanced Transportation 7
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
2000
Tram
1 V
olta
ge (V
)
Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
times106
Figure 6 Plot results trams with braking energy recovery
0 100 200 300 400 500 600 700 800 900 1000Time (s)
04
045
05
055
RESS
SO
C (p
u)
Simscape Dymola Fortran
Simscape Dymola Fortran
0 100 200 300 400 500 600 700 800 900 1000Time (s)
minus1
0
1
2
RESS
pow
er (W
)
times106
Figure 7 Plot results system equipped with storage
8 Journal of Advanced Transportation
other Modelica-compliant tool They proved to be fast withrelatively small memory occupation Nearly the same can besaid about Matlab-Simscape although characterised by lessflexibility due to the fact that it is not inspired by an open-source language platform The quality has been verified bycomparing results with those obtained using a well validatedFORTRAN based simulator
With reference to an existing case study the cost-effectiveness due to the utilisation of stationary storagesystem to enhance braking energy recovery has been clearlydemonstrated since on a high-traffic tramway with highnumber of stops payback time has been reached within onlytwo years Naturally it is of great importance in order to cor-rectly achieve energy saving for the considered application topreliminary calibrate the considered tool on actual electricalenergy consumption experimentally measured
As future direction of this work it is also possible toconsider the extension of the considered methodology toother tramlines in operation in order to investigate thepotential cost-effectiveness of similar or different energysaving solutions
Data Availability
The data used to support the findings of this study areincluded within the article
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] A Gonzalez-Gil R Palacin P Batty and J P Powell ldquoA systemsapproach to reduce urban rail energy consumptionrdquo EnergyConversion and Management vol 80 pp 509ndash524 2014
[2] A Gonzalez-Gil R Palacin and P Batty ldquoSustainable urban railsystems Strategies and technologies for optimalmanagement ofregenerative braking energyrdquo Energy Conversion and Manage-ment vol 75 pp 374ndash388 2013
[3] M C Falvo R Lamedica and A Ruvio ldquoEnergy storage appli-cation in trolley-buses lines for a sustainable urban mobilityrdquoin Proceedings of the 2012 Electrical Systems for Aircraft Railwayand Ship Propulsion ESARS 2012 Italy October 2012
[4] M C Falvo R Lamedica and A Ruvio ldquoAn environmentalsustainable transport system A trolley-buses Line for Cosenzacityrdquo in Proceedings of the 21st International Symposium onPower Electronics Electrical Drives Automation and MotionSPEEDAM 2012 pp 1479ndash1485 Italy June 2012
[5] M Berger C Lavertu I Kocar and J Mahseredjian ldquoProposalof a time-domain platform for short-circuit protection analysisin rapid transit train DC auxiliary systemsrdquo IEEE Transactionson Industry Applications vol 52 no 6 pp 5295ndash5304 2016
[6] A Ibrahem A Elrayyah Y Sozer and J A De Abreu-GarcialdquoDC railway system emulator for stray current and touchvoltage predictionrdquo IEEE Transactions on Industry Applicationsvol 53 no 1 pp 439ndash446 2017
[7] A Capasso R Lamedica andC Penna ldquoEnergy regeneration intransportation systemmethodologies for powernetworks simu-lationrdquo in Proceedings of the 4th IFACIFIPIFORS international
conference on control in transportation system Baden GermanyApril 1983
[8] A Capasso G Giannini R Lamedica and A Ruvio ldquoEco-friendly urban transport systems Comparison between energydemands of the trolleybus and tram systemsrdquo Ingegneria Fer-roviaria vol 69 no 4 pp 329ndash347 2014
[9] A Adinolfi R Lamedica C Modesto A Prudenzi and SVimercati ldquoExperimental assessment of energy saving due totrains regenerative braking in an electrified subway linerdquo IEEETransactions on Power Delivery vol 13 no 4 pp 1536ndash15411998
[10] Dymola ldquoDynamic Modeling Laboratoryrdquo documentationavailable from httpwww3dscomproductsservicescatiaproductsdymola
[11] P Fritzson Introduction toModeling and Simulation of Technicaland Physical Systems with Modelica Wiley-IEEE Press 2011
[12] M Ceraolo R Giglioli G Lutzemberger M Conte and MPasquali ldquoUse of electrochemical storage in substations toenhance energy and cost efficiency of tramwaysrdquo in Proceedingsof the 2013 AEIT Annual Conference Innovation and Scientificand Technical Culture for Development AEIT 2013 MondelloItaly October 2013
[13] S Barsali P Bolognesi M Ceraolo M Funaioli and G Lu-tzembergerCyber-Physical Modelling of Railroad Vehicle SystemusingModelica Simulation Language vol 104 Railways AjaccioCorsica France 2014
[14] ldquoMatlab-Simscaperdquo Official site httpswwwmathworkscomproductssimscapehtml
[15] A Capasso R Lamedica A Ruvio M Ceraolo and GLutzemberger ldquoModelling and simulation of electric urbantransportation systems with energy storagerdquo in Proceedings ofthe 16th International Conference on Environment and ElectricalEngineering (EEEIC rsquo16) Florence Italy June 2016
[16] ldquoCEI EN 50388 Railway applications - Power supply and rollingstockrdquo 2012
[17] M E El-Hawary and O K Wellon ldquoThe alpha-modified quasi-second order Newton-Raphson method for load flow solutionsin rectangular formrdquo IEEE Transactions on Power Apparatusand Systems vol 101 no 4 pp 854ndash866 1982
[18] A Capasso and R Lamedica Simulation of The Operation ofElectric Traction Networks CNR - Progetto finalizzato trasportiBologna Italy 1984
[19] M Ceraolo R Giglioli G Lutzemberger and A BechinildquoCost effective storage for energy saving in feeding systems oftramwaysrdquo in Proceedings of the 2014 IEEE International ElectricVehicle Conference IEVC 2014 IEEE Florence Italy December2014
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Journal of Advanced Transportation 5
Table 1 Input model parameters of the system under study
ESS parametersNo load voltage (V) 1680R0(Ω) 013
Number of ESSs 1ESS position (km) 0
Line parametersMax line voltage (V) 1800Nominal line voltage (V) 1650Min line voltage (V) 1100Number of line trunks 6
Line Resistance (Ωkm) 010 010 007007 007 007
Tram parametersFull mass (t) 92Auxiliary power adsorption (kW) 40ED max power (kW) 1242ED max traction force (kN) 96
Track parametersNumber of trams 2Number of stops 14Average distance between stops (km) 04Max speed (ms) 14Track length (km) 54
First simulations were performed without consideringbraking energy recovery As said model parameters wereslightly updated in order tomatch the results provided by thenew tools with the oldest one Figure 5 shows the plot resultsof one train driving on a portion of the considered route Inthis case the tram is not able to send its braking energy alongthe catenary As observable plot results are equivalent amongthe three tools respectively developed in Modelica Matlab-Simscape and FORTRAN
Then simulations have considered the possibility ofrecovering the braking energy In this case parameters actingon control voltage at pantograph have been tuned to performmodulation of the inlet power without overcoming voltagelimits Figure 6 shows plot results of one train under a portionof the considered profile
As noted different blending strategies according to Fig-ure 2 have been implemented Naturally these differenceshave an impact on the total energy absorbed from thenetwork The blending strategy having Vrsquo=09divide095119881max (seeFigure 2) implemented in FORTRAN results in a reductionof the energy recovered through the pantograph respectivelyof 13 and 2 with respect to the dynamical strategyused in Modelica and Matlab-Simscape that allows the fullpowertrain power to be recovered up to 119881max (ie V rsquo=119881max)Then in terms of total energy adsorption from the ESSincrement is of 3 and 06 respectively
Naturally braking energy recovery can be enhancedthrough energy storage systems installed along the route
Table 2 Lithium battery main characteristics
Nominal energy (MWh) 033Nominal capacity (Ah) 200Nominal voltage (V) 1650Number of cells in series 446Max allowed current (A) 2000Charging-discharging efficiency 09
Table 3 ESS energy consumption
ESS working day daily energy (MWh) 158ESS holiday daily energy (kWh) 111ESS Annual energy (MWh) 5386
In this way one storage system has been introduced withthe main aim of validating the tools under test also in thenew considered system configuration The lithium battery ispositioned about halfway along the tramway (ie about 38km from the terminal) whose characteristics are in Table 2
Relation among the nominal energy and nominal capac-ity is given by
119864119899= 119899119904119881119899119888119862119899 (5)
where 119899119904is the number of cells in series 119864
119899is the nominal
energy 119881119899119888is the nominal cell voltage and 119862
119899is the nominal
capacity Number of cells and nominal capacity were selectedin order to exactly reproduce through the newest tools thebattery SOC (State-of-Charge) and the power profile withrespect to the FORTRAN based one Results are shown inFigure 7
Installation of one storage system can significantly reducethe energy delivered by the electrical feeding substations(ESSs) The considered simplified case study with two trainsobviously cannot correctly evaluate the cost-effectivenessof the considered solution Thus the full analysis will bedescribed in the next section following the same approachalready considered by the authors in [12 13]
32 Energy Saving Evaluation Energy saving evaluationhas been performed by considering an experimental mea-surement campaign carried on by the authors Results aresummarized in Table 3 where the daily energy delivered fromthe electrical substation (ESS) has been measured duringone typical school working day and one typical holidayday Starting from that evaluation of the annual electricalenergy demand has been performed simply by consideringthe number of working days and holidays over the year givenby the tramline operator
Therefore the Modelica-based tool presented before hasbeen used to exactly reproduce the energy consumptionshown in Table 3 To do that different numbers of operatingtrams along the day have been considered from timetablerespectively 24 during peak hours typically concentrated inthe middle part of the day and 12 in low load hours mainlyin the early morning and in the late evening Real numberof trams in operation may be slightly different However the
6 Journal of Advanced Transportation
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
Tram
1 V
olta
ge (V
)
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
times106
Figure 5 Plot results trams without braking energy recovery
daily distribution has been chosen to exactly reproduce themeasured consumption
After simulating exactly the same level of the measuredenergy demand ie obtaining the daily consumption shownin Table 3 simulations have been repeated by including thestorage system described in Table 2 The delivered energy bythe ESS was then compared with the previous one calculatedin absence of the storage system The reduction in annualenergy consumption was about 152 moving from 5386MWh up to 4570MWh
The cost-effectiveness of the proposed solution has beeninvestigated by considering the initial cash outlay due theintroduction of the storage system with respect to theannual return of the investment due to the above-mentionedelectrical energy saving The initial cash outlay due to thestorage system and its balance of plant has been calculatedby considering a value of 500 eurokWh including cells BMSand battery packaging As discussed in detail in [19] thepresence of a DCDC converter may be required For thisa fixed cost of 60 keuro in analogue way to the experiencepresented in [19] has been used Finally the current industrialuser price of energy in Italy was evaluated considering anaverage value of 150 euroMWh In terms of maintenancecosts for the reasons already explained in [19] the storagehas been considered able to cover the whole life of theplant
Table 4 Economic benefit analysis
Storage system cost (keuro) 2261Annual energy saving (keuro) 1225NPV (keuro) 7382PBT (y) 2
Main objective of the analysis was related to the evalua-tion of the net present value (NPV) and of the payback time(PBT) for a whole life of ten years and an interest rate of 4Results are shown in Table 4
The results show that installation of a stationary storagesystem may guarantee a payback time within just two yearThese numbers are so favourable not to be affected by anypossible storage substitution during the plant life Indeed thecost-effectiveness of the proposed solution has been clearlydemonstrated
4 Conclusions
This paper has demonstrated how innovative languages andsoftware allow rapid creation of numerical models havingelectrical mechanical and control parts to be simulated
The Modelica-based model was realised through thecommercial Dymola tool However it could run inside any
Journal of Advanced Transportation 7
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
2000
Tram
1 V
olta
ge (V
)
Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
times106
Figure 6 Plot results trams with braking energy recovery
0 100 200 300 400 500 600 700 800 900 1000Time (s)
04
045
05
055
RESS
SO
C (p
u)
Simscape Dymola Fortran
Simscape Dymola Fortran
0 100 200 300 400 500 600 700 800 900 1000Time (s)
minus1
0
1
2
RESS
pow
er (W
)
times106
Figure 7 Plot results system equipped with storage
8 Journal of Advanced Transportation
other Modelica-compliant tool They proved to be fast withrelatively small memory occupation Nearly the same can besaid about Matlab-Simscape although characterised by lessflexibility due to the fact that it is not inspired by an open-source language platform The quality has been verified bycomparing results with those obtained using a well validatedFORTRAN based simulator
With reference to an existing case study the cost-effectiveness due to the utilisation of stationary storagesystem to enhance braking energy recovery has been clearlydemonstrated since on a high-traffic tramway with highnumber of stops payback time has been reached within onlytwo years Naturally it is of great importance in order to cor-rectly achieve energy saving for the considered application topreliminary calibrate the considered tool on actual electricalenergy consumption experimentally measured
As future direction of this work it is also possible toconsider the extension of the considered methodology toother tramlines in operation in order to investigate thepotential cost-effectiveness of similar or different energysaving solutions
Data Availability
The data used to support the findings of this study areincluded within the article
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] A Gonzalez-Gil R Palacin P Batty and J P Powell ldquoA systemsapproach to reduce urban rail energy consumptionrdquo EnergyConversion and Management vol 80 pp 509ndash524 2014
[2] A Gonzalez-Gil R Palacin and P Batty ldquoSustainable urban railsystems Strategies and technologies for optimalmanagement ofregenerative braking energyrdquo Energy Conversion and Manage-ment vol 75 pp 374ndash388 2013
[3] M C Falvo R Lamedica and A Ruvio ldquoEnergy storage appli-cation in trolley-buses lines for a sustainable urban mobilityrdquoin Proceedings of the 2012 Electrical Systems for Aircraft Railwayand Ship Propulsion ESARS 2012 Italy October 2012
[4] M C Falvo R Lamedica and A Ruvio ldquoAn environmentalsustainable transport system A trolley-buses Line for Cosenzacityrdquo in Proceedings of the 21st International Symposium onPower Electronics Electrical Drives Automation and MotionSPEEDAM 2012 pp 1479ndash1485 Italy June 2012
[5] M Berger C Lavertu I Kocar and J Mahseredjian ldquoProposalof a time-domain platform for short-circuit protection analysisin rapid transit train DC auxiliary systemsrdquo IEEE Transactionson Industry Applications vol 52 no 6 pp 5295ndash5304 2016
[6] A Ibrahem A Elrayyah Y Sozer and J A De Abreu-GarcialdquoDC railway system emulator for stray current and touchvoltage predictionrdquo IEEE Transactions on Industry Applicationsvol 53 no 1 pp 439ndash446 2017
[7] A Capasso R Lamedica andC Penna ldquoEnergy regeneration intransportation systemmethodologies for powernetworks simu-lationrdquo in Proceedings of the 4th IFACIFIPIFORS international
conference on control in transportation system Baden GermanyApril 1983
[8] A Capasso G Giannini R Lamedica and A Ruvio ldquoEco-friendly urban transport systems Comparison between energydemands of the trolleybus and tram systemsrdquo Ingegneria Fer-roviaria vol 69 no 4 pp 329ndash347 2014
[9] A Adinolfi R Lamedica C Modesto A Prudenzi and SVimercati ldquoExperimental assessment of energy saving due totrains regenerative braking in an electrified subway linerdquo IEEETransactions on Power Delivery vol 13 no 4 pp 1536ndash15411998
[10] Dymola ldquoDynamic Modeling Laboratoryrdquo documentationavailable from httpwww3dscomproductsservicescatiaproductsdymola
[11] P Fritzson Introduction toModeling and Simulation of Technicaland Physical Systems with Modelica Wiley-IEEE Press 2011
[12] M Ceraolo R Giglioli G Lutzemberger M Conte and MPasquali ldquoUse of electrochemical storage in substations toenhance energy and cost efficiency of tramwaysrdquo in Proceedingsof the 2013 AEIT Annual Conference Innovation and Scientificand Technical Culture for Development AEIT 2013 MondelloItaly October 2013
[13] S Barsali P Bolognesi M Ceraolo M Funaioli and G Lu-tzembergerCyber-Physical Modelling of Railroad Vehicle SystemusingModelica Simulation Language vol 104 Railways AjaccioCorsica France 2014
[14] ldquoMatlab-Simscaperdquo Official site httpswwwmathworkscomproductssimscapehtml
[15] A Capasso R Lamedica A Ruvio M Ceraolo and GLutzemberger ldquoModelling and simulation of electric urbantransportation systems with energy storagerdquo in Proceedings ofthe 16th International Conference on Environment and ElectricalEngineering (EEEIC rsquo16) Florence Italy June 2016
[16] ldquoCEI EN 50388 Railway applications - Power supply and rollingstockrdquo 2012
[17] M E El-Hawary and O K Wellon ldquoThe alpha-modified quasi-second order Newton-Raphson method for load flow solutionsin rectangular formrdquo IEEE Transactions on Power Apparatusand Systems vol 101 no 4 pp 854ndash866 1982
[18] A Capasso and R Lamedica Simulation of The Operation ofElectric Traction Networks CNR - Progetto finalizzato trasportiBologna Italy 1984
[19] M Ceraolo R Giglioli G Lutzemberger and A BechinildquoCost effective storage for energy saving in feeding systems oftramwaysrdquo in Proceedings of the 2014 IEEE International ElectricVehicle Conference IEVC 2014 IEEE Florence Italy December2014
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
6 Journal of Advanced Transportation
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
Tram
1 V
olta
ge (V
)
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
times106
Figure 5 Plot results trams without braking energy recovery
daily distribution has been chosen to exactly reproduce themeasured consumption
After simulating exactly the same level of the measuredenergy demand ie obtaining the daily consumption shownin Table 3 simulations have been repeated by including thestorage system described in Table 2 The delivered energy bythe ESS was then compared with the previous one calculatedin absence of the storage system The reduction in annualenergy consumption was about 152 moving from 5386MWh up to 4570MWh
The cost-effectiveness of the proposed solution has beeninvestigated by considering the initial cash outlay due theintroduction of the storage system with respect to theannual return of the investment due to the above-mentionedelectrical energy saving The initial cash outlay due to thestorage system and its balance of plant has been calculatedby considering a value of 500 eurokWh including cells BMSand battery packaging As discussed in detail in [19] thepresence of a DCDC converter may be required For thisa fixed cost of 60 keuro in analogue way to the experiencepresented in [19] has been used Finally the current industrialuser price of energy in Italy was evaluated considering anaverage value of 150 euroMWh In terms of maintenancecosts for the reasons already explained in [19] the storagehas been considered able to cover the whole life of theplant
Table 4 Economic benefit analysis
Storage system cost (keuro) 2261Annual energy saving (keuro) 1225NPV (keuro) 7382PBT (y) 2
Main objective of the analysis was related to the evalua-tion of the net present value (NPV) and of the payback time(PBT) for a whole life of ten years and an interest rate of 4Results are shown in Table 4
The results show that installation of a stationary storagesystem may guarantee a payback time within just two yearThese numbers are so favourable not to be affected by anypossible storage substitution during the plant life Indeed thecost-effectiveness of the proposed solution has been clearlydemonstrated
4 Conclusions
This paper has demonstrated how innovative languages andsoftware allow rapid creation of numerical models havingelectrical mechanical and control parts to be simulated
The Modelica-based model was realised through thecommercial Dymola tool However it could run inside any
Journal of Advanced Transportation 7
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
2000
Tram
1 V
olta
ge (V
)
Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
times106
Figure 6 Plot results trams with braking energy recovery
0 100 200 300 400 500 600 700 800 900 1000Time (s)
04
045
05
055
RESS
SO
C (p
u)
Simscape Dymola Fortran
Simscape Dymola Fortran
0 100 200 300 400 500 600 700 800 900 1000Time (s)
minus1
0
1
2
RESS
pow
er (W
)
times106
Figure 7 Plot results system equipped with storage
8 Journal of Advanced Transportation
other Modelica-compliant tool They proved to be fast withrelatively small memory occupation Nearly the same can besaid about Matlab-Simscape although characterised by lessflexibility due to the fact that it is not inspired by an open-source language platform The quality has been verified bycomparing results with those obtained using a well validatedFORTRAN based simulator
With reference to an existing case study the cost-effectiveness due to the utilisation of stationary storagesystem to enhance braking energy recovery has been clearlydemonstrated since on a high-traffic tramway with highnumber of stops payback time has been reached within onlytwo years Naturally it is of great importance in order to cor-rectly achieve energy saving for the considered application topreliminary calibrate the considered tool on actual electricalenergy consumption experimentally measured
As future direction of this work it is also possible toconsider the extension of the considered methodology toother tramlines in operation in order to investigate thepotential cost-effectiveness of similar or different energysaving solutions
Data Availability
The data used to support the findings of this study areincluded within the article
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] A Gonzalez-Gil R Palacin P Batty and J P Powell ldquoA systemsapproach to reduce urban rail energy consumptionrdquo EnergyConversion and Management vol 80 pp 509ndash524 2014
[2] A Gonzalez-Gil R Palacin and P Batty ldquoSustainable urban railsystems Strategies and technologies for optimalmanagement ofregenerative braking energyrdquo Energy Conversion and Manage-ment vol 75 pp 374ndash388 2013
[3] M C Falvo R Lamedica and A Ruvio ldquoEnergy storage appli-cation in trolley-buses lines for a sustainable urban mobilityrdquoin Proceedings of the 2012 Electrical Systems for Aircraft Railwayand Ship Propulsion ESARS 2012 Italy October 2012
[4] M C Falvo R Lamedica and A Ruvio ldquoAn environmentalsustainable transport system A trolley-buses Line for Cosenzacityrdquo in Proceedings of the 21st International Symposium onPower Electronics Electrical Drives Automation and MotionSPEEDAM 2012 pp 1479ndash1485 Italy June 2012
[5] M Berger C Lavertu I Kocar and J Mahseredjian ldquoProposalof a time-domain platform for short-circuit protection analysisin rapid transit train DC auxiliary systemsrdquo IEEE Transactionson Industry Applications vol 52 no 6 pp 5295ndash5304 2016
[6] A Ibrahem A Elrayyah Y Sozer and J A De Abreu-GarcialdquoDC railway system emulator for stray current and touchvoltage predictionrdquo IEEE Transactions on Industry Applicationsvol 53 no 1 pp 439ndash446 2017
[7] A Capasso R Lamedica andC Penna ldquoEnergy regeneration intransportation systemmethodologies for powernetworks simu-lationrdquo in Proceedings of the 4th IFACIFIPIFORS international
conference on control in transportation system Baden GermanyApril 1983
[8] A Capasso G Giannini R Lamedica and A Ruvio ldquoEco-friendly urban transport systems Comparison between energydemands of the trolleybus and tram systemsrdquo Ingegneria Fer-roviaria vol 69 no 4 pp 329ndash347 2014
[9] A Adinolfi R Lamedica C Modesto A Prudenzi and SVimercati ldquoExperimental assessment of energy saving due totrains regenerative braking in an electrified subway linerdquo IEEETransactions on Power Delivery vol 13 no 4 pp 1536ndash15411998
[10] Dymola ldquoDynamic Modeling Laboratoryrdquo documentationavailable from httpwww3dscomproductsservicescatiaproductsdymola
[11] P Fritzson Introduction toModeling and Simulation of Technicaland Physical Systems with Modelica Wiley-IEEE Press 2011
[12] M Ceraolo R Giglioli G Lutzemberger M Conte and MPasquali ldquoUse of electrochemical storage in substations toenhance energy and cost efficiency of tramwaysrdquo in Proceedingsof the 2013 AEIT Annual Conference Innovation and Scientificand Technical Culture for Development AEIT 2013 MondelloItaly October 2013
[13] S Barsali P Bolognesi M Ceraolo M Funaioli and G Lu-tzembergerCyber-Physical Modelling of Railroad Vehicle SystemusingModelica Simulation Language vol 104 Railways AjaccioCorsica France 2014
[14] ldquoMatlab-Simscaperdquo Official site httpswwwmathworkscomproductssimscapehtml
[15] A Capasso R Lamedica A Ruvio M Ceraolo and GLutzemberger ldquoModelling and simulation of electric urbantransportation systems with energy storagerdquo in Proceedings ofthe 16th International Conference on Environment and ElectricalEngineering (EEEIC rsquo16) Florence Italy June 2016
[16] ldquoCEI EN 50388 Railway applications - Power supply and rollingstockrdquo 2012
[17] M E El-Hawary and O K Wellon ldquoThe alpha-modified quasi-second order Newton-Raphson method for load flow solutionsin rectangular formrdquo IEEE Transactions on Power Apparatusand Systems vol 101 no 4 pp 854ndash866 1982
[18] A Capasso and R Lamedica Simulation of The Operation ofElectric Traction Networks CNR - Progetto finalizzato trasportiBologna Italy 1984
[19] M Ceraolo R Giglioli G Lutzemberger and A BechinildquoCost effective storage for energy saving in feeding systems oftramwaysrdquo in Proceedings of the 2014 IEEE International ElectricVehicle Conference IEVC 2014 IEEE Florence Italy December2014
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Journal of Advanced Transportation 7
650 700 750 800 850 900Time (s)
0
5
10
15
20
Tram
1 sp
eed
(ms
)
650 700 750 800 850 900Time (s)
1200
1400
1600
1800
2000
Tram
1 V
olta
ge (V
)
Simscape Dymola Fortran
Simscape Dymola Fortran
Simscape Dymola Fortran
650 700 750 800 850 900Time (s)
0
1
2
3
ESS
Pow
er (W
)
times106
Figure 6 Plot results trams with braking energy recovery
0 100 200 300 400 500 600 700 800 900 1000Time (s)
04
045
05
055
RESS
SO
C (p
u)
Simscape Dymola Fortran
Simscape Dymola Fortran
0 100 200 300 400 500 600 700 800 900 1000Time (s)
minus1
0
1
2
RESS
pow
er (W
)
times106
Figure 7 Plot results system equipped with storage
8 Journal of Advanced Transportation
other Modelica-compliant tool They proved to be fast withrelatively small memory occupation Nearly the same can besaid about Matlab-Simscape although characterised by lessflexibility due to the fact that it is not inspired by an open-source language platform The quality has been verified bycomparing results with those obtained using a well validatedFORTRAN based simulator
With reference to an existing case study the cost-effectiveness due to the utilisation of stationary storagesystem to enhance braking energy recovery has been clearlydemonstrated since on a high-traffic tramway with highnumber of stops payback time has been reached within onlytwo years Naturally it is of great importance in order to cor-rectly achieve energy saving for the considered application topreliminary calibrate the considered tool on actual electricalenergy consumption experimentally measured
As future direction of this work it is also possible toconsider the extension of the considered methodology toother tramlines in operation in order to investigate thepotential cost-effectiveness of similar or different energysaving solutions
Data Availability
The data used to support the findings of this study areincluded within the article
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] A Gonzalez-Gil R Palacin P Batty and J P Powell ldquoA systemsapproach to reduce urban rail energy consumptionrdquo EnergyConversion and Management vol 80 pp 509ndash524 2014
[2] A Gonzalez-Gil R Palacin and P Batty ldquoSustainable urban railsystems Strategies and technologies for optimalmanagement ofregenerative braking energyrdquo Energy Conversion and Manage-ment vol 75 pp 374ndash388 2013
[3] M C Falvo R Lamedica and A Ruvio ldquoEnergy storage appli-cation in trolley-buses lines for a sustainable urban mobilityrdquoin Proceedings of the 2012 Electrical Systems for Aircraft Railwayand Ship Propulsion ESARS 2012 Italy October 2012
[4] M C Falvo R Lamedica and A Ruvio ldquoAn environmentalsustainable transport system A trolley-buses Line for Cosenzacityrdquo in Proceedings of the 21st International Symposium onPower Electronics Electrical Drives Automation and MotionSPEEDAM 2012 pp 1479ndash1485 Italy June 2012
[5] M Berger C Lavertu I Kocar and J Mahseredjian ldquoProposalof a time-domain platform for short-circuit protection analysisin rapid transit train DC auxiliary systemsrdquo IEEE Transactionson Industry Applications vol 52 no 6 pp 5295ndash5304 2016
[6] A Ibrahem A Elrayyah Y Sozer and J A De Abreu-GarcialdquoDC railway system emulator for stray current and touchvoltage predictionrdquo IEEE Transactions on Industry Applicationsvol 53 no 1 pp 439ndash446 2017
[7] A Capasso R Lamedica andC Penna ldquoEnergy regeneration intransportation systemmethodologies for powernetworks simu-lationrdquo in Proceedings of the 4th IFACIFIPIFORS international
conference on control in transportation system Baden GermanyApril 1983
[8] A Capasso G Giannini R Lamedica and A Ruvio ldquoEco-friendly urban transport systems Comparison between energydemands of the trolleybus and tram systemsrdquo Ingegneria Fer-roviaria vol 69 no 4 pp 329ndash347 2014
[9] A Adinolfi R Lamedica C Modesto A Prudenzi and SVimercati ldquoExperimental assessment of energy saving due totrains regenerative braking in an electrified subway linerdquo IEEETransactions on Power Delivery vol 13 no 4 pp 1536ndash15411998
[10] Dymola ldquoDynamic Modeling Laboratoryrdquo documentationavailable from httpwww3dscomproductsservicescatiaproductsdymola
[11] P Fritzson Introduction toModeling and Simulation of Technicaland Physical Systems with Modelica Wiley-IEEE Press 2011
[12] M Ceraolo R Giglioli G Lutzemberger M Conte and MPasquali ldquoUse of electrochemical storage in substations toenhance energy and cost efficiency of tramwaysrdquo in Proceedingsof the 2013 AEIT Annual Conference Innovation and Scientificand Technical Culture for Development AEIT 2013 MondelloItaly October 2013
[13] S Barsali P Bolognesi M Ceraolo M Funaioli and G Lu-tzembergerCyber-Physical Modelling of Railroad Vehicle SystemusingModelica Simulation Language vol 104 Railways AjaccioCorsica France 2014
[14] ldquoMatlab-Simscaperdquo Official site httpswwwmathworkscomproductssimscapehtml
[15] A Capasso R Lamedica A Ruvio M Ceraolo and GLutzemberger ldquoModelling and simulation of electric urbantransportation systems with energy storagerdquo in Proceedings ofthe 16th International Conference on Environment and ElectricalEngineering (EEEIC rsquo16) Florence Italy June 2016
[16] ldquoCEI EN 50388 Railway applications - Power supply and rollingstockrdquo 2012
[17] M E El-Hawary and O K Wellon ldquoThe alpha-modified quasi-second order Newton-Raphson method for load flow solutionsin rectangular formrdquo IEEE Transactions on Power Apparatusand Systems vol 101 no 4 pp 854ndash866 1982
[18] A Capasso and R Lamedica Simulation of The Operation ofElectric Traction Networks CNR - Progetto finalizzato trasportiBologna Italy 1984
[19] M Ceraolo R Giglioli G Lutzemberger and A BechinildquoCost effective storage for energy saving in feeding systems oftramwaysrdquo in Proceedings of the 2014 IEEE International ElectricVehicle Conference IEVC 2014 IEEE Florence Italy December2014
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
8 Journal of Advanced Transportation
other Modelica-compliant tool They proved to be fast withrelatively small memory occupation Nearly the same can besaid about Matlab-Simscape although characterised by lessflexibility due to the fact that it is not inspired by an open-source language platform The quality has been verified bycomparing results with those obtained using a well validatedFORTRAN based simulator
With reference to an existing case study the cost-effectiveness due to the utilisation of stationary storagesystem to enhance braking energy recovery has been clearlydemonstrated since on a high-traffic tramway with highnumber of stops payback time has been reached within onlytwo years Naturally it is of great importance in order to cor-rectly achieve energy saving for the considered application topreliminary calibrate the considered tool on actual electricalenergy consumption experimentally measured
As future direction of this work it is also possible toconsider the extension of the considered methodology toother tramlines in operation in order to investigate thepotential cost-effectiveness of similar or different energysaving solutions
Data Availability
The data used to support the findings of this study areincluded within the article
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] A Gonzalez-Gil R Palacin P Batty and J P Powell ldquoA systemsapproach to reduce urban rail energy consumptionrdquo EnergyConversion and Management vol 80 pp 509ndash524 2014
[2] A Gonzalez-Gil R Palacin and P Batty ldquoSustainable urban railsystems Strategies and technologies for optimalmanagement ofregenerative braking energyrdquo Energy Conversion and Manage-ment vol 75 pp 374ndash388 2013
[3] M C Falvo R Lamedica and A Ruvio ldquoEnergy storage appli-cation in trolley-buses lines for a sustainable urban mobilityrdquoin Proceedings of the 2012 Electrical Systems for Aircraft Railwayand Ship Propulsion ESARS 2012 Italy October 2012
[4] M C Falvo R Lamedica and A Ruvio ldquoAn environmentalsustainable transport system A trolley-buses Line for Cosenzacityrdquo in Proceedings of the 21st International Symposium onPower Electronics Electrical Drives Automation and MotionSPEEDAM 2012 pp 1479ndash1485 Italy June 2012
[5] M Berger C Lavertu I Kocar and J Mahseredjian ldquoProposalof a time-domain platform for short-circuit protection analysisin rapid transit train DC auxiliary systemsrdquo IEEE Transactionson Industry Applications vol 52 no 6 pp 5295ndash5304 2016
[6] A Ibrahem A Elrayyah Y Sozer and J A De Abreu-GarcialdquoDC railway system emulator for stray current and touchvoltage predictionrdquo IEEE Transactions on Industry Applicationsvol 53 no 1 pp 439ndash446 2017
[7] A Capasso R Lamedica andC Penna ldquoEnergy regeneration intransportation systemmethodologies for powernetworks simu-lationrdquo in Proceedings of the 4th IFACIFIPIFORS international
conference on control in transportation system Baden GermanyApril 1983
[8] A Capasso G Giannini R Lamedica and A Ruvio ldquoEco-friendly urban transport systems Comparison between energydemands of the trolleybus and tram systemsrdquo Ingegneria Fer-roviaria vol 69 no 4 pp 329ndash347 2014
[9] A Adinolfi R Lamedica C Modesto A Prudenzi and SVimercati ldquoExperimental assessment of energy saving due totrains regenerative braking in an electrified subway linerdquo IEEETransactions on Power Delivery vol 13 no 4 pp 1536ndash15411998
[10] Dymola ldquoDynamic Modeling Laboratoryrdquo documentationavailable from httpwww3dscomproductsservicescatiaproductsdymola
[11] P Fritzson Introduction toModeling and Simulation of Technicaland Physical Systems with Modelica Wiley-IEEE Press 2011
[12] M Ceraolo R Giglioli G Lutzemberger M Conte and MPasquali ldquoUse of electrochemical storage in substations toenhance energy and cost efficiency of tramwaysrdquo in Proceedingsof the 2013 AEIT Annual Conference Innovation and Scientificand Technical Culture for Development AEIT 2013 MondelloItaly October 2013
[13] S Barsali P Bolognesi M Ceraolo M Funaioli and G Lu-tzembergerCyber-Physical Modelling of Railroad Vehicle SystemusingModelica Simulation Language vol 104 Railways AjaccioCorsica France 2014
[14] ldquoMatlab-Simscaperdquo Official site httpswwwmathworkscomproductssimscapehtml
[15] A Capasso R Lamedica A Ruvio M Ceraolo and GLutzemberger ldquoModelling and simulation of electric urbantransportation systems with energy storagerdquo in Proceedings ofthe 16th International Conference on Environment and ElectricalEngineering (EEEIC rsquo16) Florence Italy June 2016
[16] ldquoCEI EN 50388 Railway applications - Power supply and rollingstockrdquo 2012
[17] M E El-Hawary and O K Wellon ldquoThe alpha-modified quasi-second order Newton-Raphson method for load flow solutionsin rectangular formrdquo IEEE Transactions on Power Apparatusand Systems vol 101 no 4 pp 854ndash866 1982
[18] A Capasso and R Lamedica Simulation of The Operation ofElectric Traction Networks CNR - Progetto finalizzato trasportiBologna Italy 1984
[19] M Ceraolo R Giglioli G Lutzemberger and A BechinildquoCost effective storage for energy saving in feeding systems oftramwaysrdquo in Proceedings of the 2014 IEEE International ElectricVehicle Conference IEVC 2014 IEEE Florence Italy December2014
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom