di dual fuel

7/28/2019 DI Dual Fuel

1/13

400 Commonwealth Drive, Warrendale, PA 15096-0001 U.S.A. Tel: (724) 776-4841 Fax: (724) 776-5760

SAE TECHNICALPAPER SERIES 2000-01-0286

Development of a Simulation Model for DirectInjection Dual Fuel Diesel-Natural Gas Engines

D. T. Hountalas and R. G. PapagiannakisNational Technical University of Athens

Reprinted From: Vehicle and Engine Systems Models(SP1527)

SAE 2000 World CongressDetroit, Michigan

March 6-9, 2000


2/13

The appearance of this ISSN code at the bottom of this page indicates SAEs consent that copies of thepaper may be made for personal or internal use of specific clients. This consent is given on the condition,however, that the copier pay a $7.00 per article copy fee through the Copyright Clearance Center, Inc.Operations Center, 222 Rosewood Drive, Danvers, MA 01923 for copying beyond that permitted by Sec-tions 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying such ascopying for general distribution, for advertising or promotional purposes, for creating new collective works,or for resale.

SAE routinely stocks printed papers for a period of three years following date of publication. Direct yourorders to SAE Customer Sales and Satisfaction Department.

Quantity reprint rates can be obtained from the Customer Sales and Satisfaction Department.

To request permission to reprint a technical paper or permission to use copyrighted SAE publications inother works, contact the SAE Publications Group.

No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written

permission of the publisher.

ISSN 0148-7191Copyright 2000 Society of Automotive Engineers, Inc.

Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solely

responsible for the content of the paper. A process is available by which discussions will be printed with the paper if it is published in

SAE Transactions. For permission to publish this paper in full or in part, contact the SAE Publications Group.

Persons wishing to submit papers to be considered for presentation or publication through SAE should send the manuscript or a 300

word abstract of a proposed manuscript to: Secretary, Engineering Meetings Board, SAE.

Printed in USA

All SAE papers, standards, and selected

books are abstracted and indexed in theGlobal Mobility Database


3/131

2000-01-0286

Development of a Simulation Model for Direct Injection Dual

Fuel Diesel-Natural Gas Engines

D. T. Hountalas and R. G. PapagiannakisNational Technical University of Athens

Copyright 2000 Society of Automotive Engineers, Inc.

ABSTRACT

During the last years a great deal of effort has beenmade for the reduction of pollutant emissions from directinjection Diesel Engines. Towards these efforts engineershave proposed various solutions, one of which is the use

of gaseous fuels as a supplement for liquid diesel fuel.These engines are referred to as dual combustionengines i.e. they use conventional diesel fuel andgaseous fuel as well. The ignition of the gaseous fuel isaccomplished through the liquid fuel, which is auto-ignited in the same way as in common diesel engines.One of the fuels used is natural gas, which has arelatively high auto-ignition temperature. This isextremely important since the CR of most conventionaldiesel engines can be maintained. In these engines thereleased energy is produced partially from thecombustion of natural gas and from the combustion ofliquid diesel fuel. The aim for the usage of dual-fuel

combustion systems is mainly to reduce the particulateemissions (soot) by replacing diesel fuel with natural gaspartially or entirely. For this reason in the present workare given preliminary results of a theoretical investigationusing a simulation model developed for dual fuel engines.The model is a two-zone combustion one, taking intoaccount, on a zonal basis, details of diesel fuel sprayformation and the mixing with the surrounding gas, whichis a mixture of air and natural gas. The main differencefrom a conventional diesel engine combustion system isthat the natural gas is already mixed and ready forcombustion. The natural gas burning initiates after theignition of the diesel fuel and its rate depends on the rate

of entrainment of surrounding gas inside the fuel jetformed. A soot model has been used to estimate theformation of soot while a detailed equilibrium model hasbeen used to determine the concentration of chemicalspecies. For nitric oxide the extended Zeldovichmechanism is used. The model is applied on a singlecylinder test engine located at the authors laboratory atvarious operating conditions of the engine. The amountof liquid fuel supplemented by natural gas has beenvaried and its affect on engine performance andemissions has been examined. From these preliminaryresults it is revealed a serious effect on the heat release

rate inside the engine cylinder and a reduction oparticulate emissions when compared to experimentadata obtained from the engine using diesel fuel only.

INTRODUCTION

The use of alternative gaseous fuels in engines for theproduction of power has been increasing worldwide. Thishas been prompted by the cleaner nature of theircombustion compared to conventional liquid fuels as welas their relative increased availability at attractive pricesDiesel engines can be made to operate on gaseous fuelsefficiently [1-4]1*. These engines usually have thegaseous fuel mixed with air at the start of thecompression stroke. This mixture does not auto-ignitedue to its higher self ignition temperature. The resultingmixture after the compression stroke is ignited throughthe ignition of the amount of diesel fuel injected. Diesefuel can auto-ignite readily creating ignition sources for

the surrounding gaseous fuel mixture. Most current duafuel engines are made to operate either on gaseous fuelswith diesel ignition or only on liquid fuel injection asnormal diesel engines [1-4].

A computer based mathematical model can provide anadequate way for describing details of the complicatedmixing, combustion and pollutants emission processes inthese engines. The present contribution describes amodel that simulates dual-fuel combustion using a twozone approach to simulate the combustion of premixedgas/air charge.

Its main purpose is to describe the operation of existing

diesel engines, where part of the liquid fuel is replaced bygaseous fuel. These engines are usually referred to asfumigated ones and the replacement of liquid fuel withgaseous aims mainly to the reduction of soot emissions.

Furthermore due to the partial replacement of liquid fuewith gaseous one no specific modifications are requiredand thus the technique can be applied on existingengines.

1. Numbers in brackets designate references at the end of

the paper.


4/132

The main aim of the present model is to estimate, apartfrom power and efficiency, the concentration of sootemissions and nitrogen oxides at the engine exhaust(NO), by replacing diesel fuel with natural gas partially atvarious amounts ranging from 0-40%. For this reasonpreliminary results from the modeling are provided forvarious engine operating conditions. To validate themodel under normal diesel operations and to comparethe findings under dual-fuel operation an extendedexperimental is conducted on a single cylinder testengine. From the analysis of results obtained it isrevealed that the simulation model developed predictsadequately engine performance and pollutants undernormal diesel engine operations. Furthermore comparingthe findings for dual-fuel operation with the standarddiesel one, a serious effect of the gaseous fuel isobserved on engine performance and soot, NOemissions. To validate the findings of the dual-fuelmodeling an experimental investigation is currently underway and results will be presented in the near future.

GENERAL DESCRIPTION OF THE MODEL

The cylinder charge at the start of compression stroke isa homogeneous mixture of gaseous fuel and air, that hasdifferent thermodynamic properties from the air charge innormal diesel engines. We replace a part of liquid fuelwith gaseous fuel in order to have the same poweroutput.

For combustion a two-zone model is considered [5]. Ineach zone there is a uniformity in space of pressure,temperature and composition at each instant of time,neglecting heat transfer between the zones. The firstzone consists of air/gaseous fuel (unburned zone) and

the second zone consists of combustion productsunburned mixture and evaporated liquid fuel (burningzone). These zones are separated by the area of theconical jet, which is formed during the injection of thediesel fuel [6]. Steady state jet theory, including wallimpingement, is used to describe the fuel-air mixingprocess. Ignition of the gaseous fuel, which is alreadymixed and ready for combustion, occurs as a result of thediesel jet auto-ignition after the ignition delay period ofthe diesel fuel [4]. Due to the lean nature of thesurrounding mixture no flame is considered, i.e. thecombustion rate of gaseous fuel depends on itsentrainment rate inside the burning zone. Of course this

consideration will have to be modified in the case ofnormal dual-fuel engines where the liquid fuel is a verysmall percentage of the total and used for the ignition ofthe main charge.

For the heat transfer calculation between the charge andcylinder wall, the Annand formula is employed [7]. Forprediction of soot emissions, the amount of net soot is

calculated considering the difference between the ratesof soot formation and soot oxidation inside the burningzone [6-14].

Dissociation of combustion products is taken into accounby incorporating the Vickland et al [15] method includingeleven species. For the formation of nitric oxides theextended Zeldovich chain reaction mechanism isconsidered [6-10,13-16,19].

The liquid fuel used is dodecane (C12

H26

) representingwidely the commercial diesel fuel and the gas fuel used isa mixture of methane (CH4) and propane (C3H8) inproportion (90%-10%) in (v/v), which is a typicarepresentative of natural gas fuel.

MATHEMATICAL TREATMENT

CONSERVATION OF ENERGY As already statedinside the combustion chamber there are considered twozones, a burning one consisting of air, evaporated fuelgaseous fuel plus burned products and an unburned oneconsisting of air and gaseous fuel. Each zone posses its

own temperature and composition. The pressure is thesame for the both zones. The first low equation for eachzone may be written as [6,9]

(1

where dmfbhf is the total enthalpy addition to theburning zone from the liquid diesel fuel and dmbuhbu isthe total enthalpy addition to the burning zone from theunburned zone due the entrainment rate of unburnedmixture. The internal energies and enthalpies in the

above equation considered are total, so that the heat ofcombustion is taken into account implicitly. Thetemperature, pressure and volume of each zone isobtained from the integrating of a set of three ordinaryfirst order differential equations with unknowns Tb, Tuand P. These equations have been derived after somemathematical elaboration from the first law and perfecgas state equations for both zones and the volumebalance. The system of the three equations has asfollows

(2

(3

( ) ffbbubu

u,bu,bu,b

hdmhdm

dVPdQdU

+

+=

( )

puu

uuuuuuuuuu

Cm

/dmudmhdPVdmTRdQdT ++=

pbb

bspeciesk

kkbbbfevfev

ububbbbbbbb

b

C/m

udymdmudmh

dmhdPVdRTmdmTRdQ

dT

+

++

= =


5/133

(4)

The values of Tb, Tu and P must satisfy thecontinuity equation and volume balance as well as, that

(5)

HEAT TRANSFER MODEL At each time step, wecalculate the total area of the cylinder as

(6)

Once the area is known and employing the Annandexpression [14], we have

(7)

where ,b,c are constants and is the thermalconductivity. The only problem existing is the applicationof the above equation during the combustion stroke,

since the burning zone is not entirely in contact with thecylinder wall surfaces. [8] For this reason a bulk averagetemperature of both zones is used as follows

(8)

The thermal conductivity and dynamic viscosity arecalculated using this bulk average temperature. The totalheat exchange rate is then distributed among the twozones according to their mass, temperature and specific

heat capacity as follows

(9)

ESTIMATION OF INJECTION RATE The volume flowrate of diesel fuel and its injection rate are calculatedusing the following equations :

(10

where Pnoz is the pressure difference across theinjector hole as

(11

where f is the liquid fuel density, dnoz is the diameteof the nozzle and CDinj is the nozzle dischargecoefficient. The fuel line pressure Pfl is an input to theprogram together with the dynamic injection timing. Thetotal mass of injected liquid fuel per cylinder cycle isdetermined from the integration of EQ. (10) [16].

MASS ENTRAINMENT RATE After initiation of fueinjection the burning zone begins to form and penetratesinto the combustion chamber. The present model, whilebased on simplified relations, represents the abovemechanism with satisfying accuracy. For jet penetration

the proposal of Hiroyasu et al [13] has been adopted.

Inside the jet we assume uniformity of densitytemperature and composition. The resulting fuel jet isconsidered to have the shape of a cone. The volume ofunburned mixture entrained by the jet is derived from itsvolume change. The jet is considered to be steady andconsisting of a homogenous gas mixture and itspenetration length is given as [10,13]

(12

From the previous equation we obtain, after derivationwith respect to time t, the jet velocity as

(13

If S2 and S1 represent the jet tip position at timest2=t1+dt and t1 respectively, then the massentrainment rate before impingement is given by

(14

where is the jet cone angle given by

(15

In most models the assumption is made that the burningzone after impingement follows a path parallel to thecylinders walls. In the present work the existing weltested wall jet theory of Glauert [17] is used to

[ ]

++

+++

++

=

=

b

pb

bu

pu

ucyl

bspeciesk

kkbbbfevfev

ububbbbbbb

pb

b

bbbbbbuuu

uuuuuuuu

pu

u

VC

RV

C

RV

/

]udymdmudmh

dmhdRTmdmTRdQ[C

R

PdVdRTmdmTRdmTR

dmudmhdmTRdQC

R

dP

cylbu

totalbu

VVV

mmm

=+=+

Head.cyllinerpistontot AAAA ++=

( ) ( )

+= 4w

4gwg

b

tot TTcTTD

ReaAdQ

=

=

=

u,bi

vii

u,bi

ivii

gCm

TCm

T

=

=

u,bi

ivii

iviiu,b

TCm

TCmdQdQ

fff

f

noz2noz

Dinjf

qm

P2

4

dCq

!!

!

=

=

cylflnoz PPP =

tdP

95.2S inj

25.0

u

noz

=

tdP

475.1U inj

25.0

u

noz

=

( )31322u SStan3

dm

=

o

u2 001.075.0tan

=


6/134

determine the burning zone history after wallimpingement upon the cylinder walls. At this point theunburned mixture is entrained only from the wall jet front.

The entrainment rate into the wall jet is estimated from itsvolume change as follows,

(16)

where to is the transient time for the change of the freejet into a wall jet, Rc is the cylinder radius and tw is thewall travel time.

IGNITION DELAY MODEL Before the injected liquidfuel ignites, it undergoes a delay period. The ignitiondelay period is estimated from the following correlation[18]

(17)

Ignition initiates when the previous integral becomesequal to one, the integration having started from theinitiation of injection. At this point we must state that dueto its lower cetane number, auto-ignition of the gaseousfuel is avoided, since in the opposite case seriousproblems would arise [11]. Thus the liquid fuel is theignition source for the gaseous fuel.

BURNING MODEL FOR THE LIQUID FUEL The semi-empirical model of Whitehouse-Way is used forcalculating the rate of combustion of Diesel fuel. Thepreparation rate of injected fuel is given by [5]

(18)

where mfi is the total amount of fuel injected up to theconsidered time, mfupr is the total unprepared fuel,PO2 is the partial pressure of available oxygen andk,x,m are constants.

The burning rate of the diesel fuel is effected by anArrhenious type equation. This burning rate is given by

(19)

where mfav is the total amount of unburned fuel, maavis the total amount of unburned air and st is thestoichiometric equivalence ratio of the liquid fuel.

At the beginning of the combustion period, combustion iscontrolled by the reaction rate. In a short time, after theprepared fuel during the delay period is consumed,combustion is controlled by the preparation rate.

ESTIMATION OF GASEOUS FUEL BURNING RATE As already stated in the present work, we replace a parof liquid fuel with gaseous fuel in order to have the samepower output. The working media at the start of thecompression stroke is a mixture of air and gaseous fuelThus the temperature level of the mixture during thecompression stroke is very important for establishingwhether auto-ignition takes place. Experiments incombustion bombs have shown that auto-ignition ogaseous fuel under diesel-like conditions requires highetemperatures from those achieved during thecompression stroke [1,2,3], for this reason auto-ignition isavoided.

During the combustion stroke the mass of the gaseousfuel, entrained inside the burning zone, is proportionathe entrainment rate of unburned mass. The combustionrate of gaseous fuel is controlled by the entrainment rateand the local conditions inside the burning zone. Thereaction rate is determined using an Arrhenious typeequation as follows, [10]

(20

where mgifav is the available mass of gaseous fuelmO2av is the available mass of oxygen in the burningzone, "Egif" is the activation energy of gaseous fuel andA,z1,z2 are constants. The total rate of heat release isobviously the sum of the two rates, the liquid fuel hearelease and the gaseous one.

CHEMICAL SPECIES FORMULATION Combustionproducts are defined by dissociation considerations. Fothe C-H-O system the complete chemical equilibriumscheme proposed by Vickland et al. (1962) is used [15]For the combustion zone, given its volume, temperaturemass of fuel burned and mass of air entrained, theconcentration of each one of the eleven species can becalculated by solving a system of 11 equationscontaining: 4 atom balance equations (one for eachelement C,H,O,N) and 7 equilibrium equations. Aparfrom soot, at any instant of time, the following gasspecies are considered to be present inside each burnedregion, all in chemical equilibrium,

(1) H2O 2) H2 (3) OH (4) H(5) N2 (6) NO (7) N (8) CO2(9) CO (10) O2 (11) O

where the species are referred to by the number inparenthesis, in front of their name, and expressed bymeans of kmole fraction X. The equilibriumconcentrations of the eleven species can be described bythe following seven equilibrium reactions (B=X1/X2):

5.0o

5.11w

5.12w22

cut5.4

tttanUR

3dm

=

dt

T

5500expPa

1S

t

0

b

757.0del

pr

=

mO

xfupr

x1fifpr 2

Pmmkdm =

( )staavfav

b

757.0fb

m,mmin

T

5500expP'kdm

=

=

=

bm

fg

z

b

avO

z

b

favg

HC,CHi

fbg

TR

Eexp

m

m

m

mdm

i

2

2

1

i

834

i


7/135

(21)

where Kp are the reactions equilibrium constants.

The solution of the above equations is obtained byincorporating the method proposed by Vickland et al(1962), with some slight modifications to ensure fastconversion.

NITRIC OXIDE CHEMICAL KINETICS Since theformation of nitric oxides is a kinetically controlledmechanism, the extended Zeldovich chain reactionmechanism is used. The three following reactions areconsidered

(22)

After some transformations the kmoles of NO areexpressed as follows:

(23)

where (NO) denotes concentration, = (NO)/(NO)e andsubscript "e" denotes equilibrium. "V" is the burned gasvolume and RI (i=1,2,3) is the one way equilibrium rate,for the "i" reaction, defined as follows:

(24)

and kif is the forward reaction rate constant for the ireaction.

SOOT FORMATION MODEL The modeling of sootformation is a very difficult task in common internalcombustion engine simulation models. For this reasonresearchers usually use semi-empirical models which

have been derived as correlation from the analysis ofexperimental data. In the present work a semi-empiricamodel, that has been widely tested, is used to predict therate of soot formation. [6,8-10,13]

The soot formation and oxidation rates are givenrespectively by,

(25

(26

where ms is the net soot formed, PO2 is the partiapressure of oxygen and Af and Ab are constants.

The net soot formation rate is then obtained from theexpression:

(27

EXPERIMENTAL FACILITIES AND PROCEDURE

To calibrate the present model an experimentainvestigation has been conducted on a single cylinderLister LV1, direct injection diesel engine located at theauthors laboratory. The results of this investigation areused to calibrate and evaluate the model under normadiesel fuel operation. These results are also used asbasis to evaluate the findings when using gaseous fuel asa supplement for liquid fuel. The technical data of theengine are given in Table 1. This is a naturally aspiratedair-cooled, four-stroke engine, with a bowl-in-pistoncombustion chamber. The normal speed range is 1000

3000 rpm. A three hole injector nozzle (each hole havinga diameter of 0.24mm) is located in the middle of thecombustion chamber head. The injector nozzle openingpressure is 190 bar. The main parts of the test installationused are:

Lister LV1 diesel engine.

Heenan & Froude hydraulic dynamometer.

Tank and flow meter for diesel fuel.

TDC marker (magnetic pick-up) and rpm indicator.

K-type thermocouples for measuring thetemperatures of the exhaust gas and engine oil.

Kistler 6001 miniature piezoelectric transducer fomeasuring the pressure in the cylinder, flushmounted to the cylinder head and carefullycalibrated.

Kistler 7063 piezoelectric transducer for measuringfuel-line pressure before the injector.

Kistler 5007 charge amplifiers.

( )

( )567P222

896P222

235P22

2104P222

573p2

10112p2

241p2

XBXPK,NOHN2

1OH

XBXK,COOHHCO

XBXPK,H

2

1OHOH

BPXK,OH2OH2

XXPK,NN2

1

XXPK,OO2

1

XXPK,HH2

1

=++

=++

=+

=+

=

=

=

)(

10f3

T/31256f22

10f12

102.4k,HNOOHN

eT104.6k,ONOON

106.1k,ONNON

=++

=++

=++

( )[ ] ( )

+

+

=

32

1

12

RR

R1

R12

dt

VNOd

V

1

( ) ( )

( ) ( )

( ) ( )eef33

e2ef22

eef11

OHNkR

ONkR

NONOkR

=

=

=

=

bm

sf5.0favfsf

TR

EexpPmAdm

=

bm

sb8.1O

sbsbTR

EexpP

P

PmAdm 2

sbsfs dmdmdm =


8/136

The output signals from the magnetic pick-up andpiezoelectric transducers are connected to the input of aKEITHLEY DAS-1801ST A/D board, installed on a IBMcompatible Pentium II PC.

The exhaust gas analysis system consists of a group ofanalyzers for measuring all the main gaseous pollutantstogether with soot concentration. The nitric oxideconcentration (NO) was measured by a standard Signalchemiluminiscent analyzer having a NO2 to NO converterswitch and fitted with a heated line, while the sootconcentration was measured using a Bosch RTT100analyzer.

The series of experiments conducted for this preliminarystudy, involved as already mentioned only the diesel fuel(no gas was added) in a variety of engine operatingconditions. The following discussion refers to the resultsfor an engine speed of 1500 rpm and four different loads,namely 20%, 40%, 60% and 80% of full engine load.Currently efforts are given to conduct experiments usinggaseous fuel as a supplement for liquid fuel taking intoaccount the observations obtained from the modeling.These results will be used to evaluate the findings of thenewly developed dual-fuel model.

RESULTS AND DISCUSSION

Figures 1 to 4 present the comparison betweentheoretical and experimental pressure and heat releasetraces in the case of the 100% Diesel fuel, for 1500 rpmengine speed and four different loads namely 20, 40, 60

and 80% of full load respectively. It can be observed thatthe agreement in all cases is good, a fact that ispromising for the utilization of the model to predict engineperformance for dual fuel engines. It must be stated thathe values of the two-zone models constants are held thesame for the entire range of operating parametersvariation and for all fuel mixtures considered in thepresent study. Results for the experimental heat releaserate curves in the above figures were obtained from theanalysis of the corresponding experimental cylindepressure traces using a diagnostic code [20].

Figure 1. Comparison between experimental andcalculated pressures and heat release tracesat 1500 rpm engine speed and 20% loadunder 100% diesel fuel operation.


Table 1. Engine basic design data, Lister LV1-Dieselhigh speed engine

Bore 85.73mm

Stroke 82.55mm

Connecting Rod Length 148.59mm

Compression Ratio 18

Cylinder Dead Volume 28.03cm3

Inlet Valve Opening 15CA before TDC

Inlet Valve Closure 41CA after BDC

Exhaust Valve Opening 41CA before BDC

Exhaust Valve Closure 15CA after TDC

Inlet Valve Diameter 34.5mm

Exhaust Valve Diameter 31.5mm

Static Injection Timing 28CA before TDC

100 120 14 0 1 60 18 0 20 0 220 2 40

Crank Angle (deg CA)

0

10

20

30

40

50

60

70

80

P

ressure

(bar)

0

50

1 00

1 50

2 00

2 50

H

ea

tR

elease

R

ate

J/de

20% Load-1500 rpmFuel:Diesel

Calculated

Exp. Cyl. Pressure

Exp. Heat Release

100 120 14 0 1 60 18 0 20 0 220 2 40


0

10

20

30

40

50

60

70

80

P

ressure

(bar)

0

50

1 00

1 50

2 00

2 50

H

eatR

elease

R

ate

J/d

e

40% Load-1500 rpmFuel:Diesel

Calculated

Exp. Cyl. Pressure

Exp. Heat Release


9/137



The variation of maximum cylinder pressure with load byvarious percentages of gaseous fuel is given in Figure 5.As shown the maximum combustion pressure is affectedby the presence of the gaseous fuel in the correspondingmixtures. Combustion of the diesel-gas fuel mixtureresults to higher rates of heat release and this in turn

results to higher combustion pressures compared to purediesel fuel under the same conditions. Differencesbecome higher as engine load and percentage of liquidfuel replacement by gaseous fuel increase.

An explanation for the previous, is given by Figure 6,which presents the corresponding heat release ratediagrams for the two extreme mixtures considered at 40and 80% load.

Figure 5. Maximum combustion pressure vs load forvarious percentages of Diesel/Gaseous fuel.

It is observed that during the premixed-controlledcombustion phase of the liquid fuel, a sudden increase inheat release rate occurs, at higher load conditions. This

is due to the sharp rate of combustion caused bypresence of the gaseous fuel, which is already premixedwith air and ready for combustion.

Figure 6. Heat release rate at 1500 rpm and 80%, 40%load for 100/0 and 60/40 Diesel/GaseousFuel operation

Figures 7 to 9 present the comparison betweentheoretical pressure traces in the case of 100%, 90%80%, 70%, and 60% Diesel fuel for 1500 rpm and fo

1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0


0

1 0

2 0

3 0

4 0

5 0

6 0

7 0

8 0

P

ressure

(bar)

0

50

10 0

15 0

20 0

25 0

HeatR

elease

Rate

(J/deg)60% Load -1500 rpm

Fuel :Diese l

C alcu lated

E x p . Cy l . P r es s u r e

E x p . H ea t R e leas e

1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0


0

1 0

2 0

3 0

4 0

5 0

6 0

7 0

8 0

Pressure

(bar)

0

50

10 0

15 0

20 0

25 0

HeatRelease

Rate

(J/deg

)80% Load -1500 rpm

Fuel :Diese l

C alcu lated

E x p . Cy l . P r es s u r e

E x p . H ea t R e leas e

1 2 3 4 5 6

B.M.E.P. (bar)

50

60

70

80

90

M

axim

um

Pressure

(bar) 100/0-D iese l /G as

90 /10-D iese l /G as




1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0


0

5 0

10 0

15 0

20 0

25 0

30 0

TotalHeatRelease

Rate

(J/deg)

0

50

10 0

15 0

20 0

40% l oad -1500 rpm


60/4 0-D iese l /G as

80% l oad -1500 rpm


60/40 -D iese l /G as


10/138

three different loads. It is observed that under part loadconditions the presence of gaseous fuel affect slightly thevalues of the pressure. This is due to the fact that duringthe premixed-controlled combustion phase the heatrelease rate at part load when using a mixture of Diesel-Gas fuel is similar to the one when using 100% dieselfuel. Under high load conditions the increase of thepercentage of gaseous fuel replacing the diesel fuel,leads to a sharper rate of heat release during thepremixed-controlled phase, as already shown in Figure 6.

Figure 7. Comparison between calculated Pressuretraces for various percentages of Diesel/Gaseous fuel at 1500 rpm and 40%load.


Figure 10 presents the variation of brake specific fuelconsumption with engine load. It is observed that in thecase of 100% diesel fuel operation the model predictsaccurately the measured bsfc values for all loadsexamined.


It is observed that the increment of gas percentage in thefuel results in an analogue increment of engine efficiencywhich becomes higher as load increases.

Figure 10. Brake specific fuel consumption at variousengine loads for various percentages ofDiesel/Gaseous fuel.

The variation of Nitric Oxide concentration with engineload is presented in Figure 11. Comparing the calculatedvalues of NO with the measured ones at 100% diesel fue

operation, it is revealed a good coincidence. The modeover-predicts slightly absolute values, which is normal foa two-zone model, but it manages to predict the trendwith load. This enables us to use it for NO predictionunder dual fuel operation. The formation of Nitric Oxide inthe fuel jet is directly related to the local oxygenconcentration and the gas temperature. Formation oNitric Oxide takes place mainly during the premixed andinitial part of mixing-controlled combustion, where high

1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0

C r a n k A n g l e (d e g C A )

0

10

20

30

40

50

60

70

80

Pressure

(bar)

40% Load -1500 rpm

100/0-Dies e l /G as

90 /10-Dies e l /G as




1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0


0

1 0

2 0

3 0

40

5 0

6 0

7 0

80

Pressure

(bar)

60% Load -1500 rpm


90/1 0-D iesel /G as




1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0 2 4 0


0

1 0

2 0

3 0

4 0

5 0

6 0

7 0

8 0

90

P

ressure

(bar)

80% Load -1500 rpm






1 2 3 4 5 6

B.M.E.P. (bar)

90

1 20

1 50

18 0

2 10

2 4 0

2 70

3 0 0

B

.S.F.C.

(gr/PS/hr)

1 0 0 / 0 -Di es el/ Gas

9 0 / 1 0 -Di es el/ Gas

8 0 / 2 0 -Di es el/ Gas

7 0 / 3 0 -Di es el/ Gas

6 0 / 4 0 -Di es el/ Gas


11/139

local temperatures exist. At part load conditions the NOconcentrations differ slightly for various percentages ofDiesel/Gas mixtures due to the previously mentionedeffect on the heat release rate. As load increases nitricoxide concentration increase with the percentage ofliquid fuel replaced by the gaseous one.

Figure 11. Nitric Oxide emissions at 1500 rpm andvarious loads for various percentages ofDiesel/Gaseous fuel

Figure 12. Soot emissions at 1500 rpm and variousloads for various percentages of Diesel/Gaseous fuel

Figure 12 presents the variation of the soot concentration

in the engine exhaust with load for various proportions ofDiesel/gas mixture. Observing Figure 12 we can see thatthe current model predicts adequately the experimentallymeasured soot emissions at 100% diesel fuel operation.This is encouraging and makes the results obtainedunder dual fuel operation more reliable. A difference isobserved at 60% load, which is due to the differentbehaviour of the engine fuel injection system, asobserved from the measured values of fuel injectionpressure. As this percentage of gaseous/liquid fuelincrease, the soot concentration is decreased since less

liquid fuel is injected and thus less soot is formed. Alsodue to the higher temperatures existing at highpercentages of liquid fuel replacement, the soot oxidationrate is higher contributing to a further decrease of emittedsoot. The decrease of soot is more intense at high loads.

CONCLUSIONS

Under the present work a new model has been

developed to simulate the operation of dual-fuel dieseengines for the prediction of performance and pollutanemissions. This preliminary investigation focuses onexisting diesel engines where a part of the liquid fuel isreplaced with a gaseous one equivalent to natural gasThe main purpose is to examine the effect of liquid fuepercentage replaced by gaseous fuel on performanceand mainly soot and nitric oxide emissions.

For this reason a two-zone model has been developedcapable of operating with various percentages of liquidand gaseous fuel. Combustion is initiated by the ignitionof the liquid fuel, while the burning rate of gaseous fuel iscontrolled by its entrainment rate into the burning zoneDue to the lean nature of the surrounding unburned air-gaseous fuel mixture no flame front is considered insidethe unburned zone. To validate the model before using ito predict engine performance and pollutants emissionsunder dual-fuel operation an extended experimentainvestigation has been conducted on a high speed Dsimple cylinder test engine located at our laboratoryMeasurements have been taken for both performanceand emissions at various operating conditions usingdiesel fuel only. These are used for model validation andalso serve as comparison between normal diesel fueoperation and operation when using various percentages

of liquid and gaseous fuel.Comparing calculated and measured values undenormal diesel operation a good coincidence is observedfor both performance and pollutant emissions. Thisenables us to use the model to predict engine behaviourunder dual-fuel operation. From the analysis ocomputational data it is revealed that dual-fuel operationresults to higher combustion pressures. These pressuresincrease with increasing percentages of gaseous/liquidfuel. The effect at high engine loads is more intenseConcerning engine efficiency it is revealed in general thathe replacement of liquid fuel with gaseous results to animprovement of engine efficiency. This improvement ismore intense at higher engine loads and higher values ofthe gaseous/liquid fuel ratio.

As far as pollutant emissions are concerned the use ofgaseous fuel has a negative effect on NO emissions anda positive on soot. Specifically an increase of NOemissions is observed which is serious at high engineloads and gaseous/liquid fuel ratios. On the other handsoot is seriously decreased when using gaseous fuel asa supplement for liquid fuel. Again this effect is stronge

1 2 3 4 5 6

B.M.E.P. (bar)

3 00

4 50

60 0

7 50

9 00

1 0 5 0

1 2 0 0

1 3 5 0

1 5 0 0

1 6 5 0

1 8 0 0

1 9 5 0

NitricOxide

(ppm

)

100/0-Dies e l /G as





E xp . D i e se l

1 2 3 4 5 6

B.M.E.P. (bar)

0 . 0 0

0 . 0 2

0 . 0 4

0 . 0 6

0 . 0 8

0 . 1 0

0 . 1 2

0 . 1 4

0 . 1 6

0 . 1 8

0 . 2 0

S

oot(m

g/lt)

100/0-D iese l /G as





E xp . D i e se l


12/1310

at higher engine loads and gaseous/liquid fuel ratios. Butthe most important is that when using gaseous fuel wealways have a reduction of soot emission regardless ofthe engine operating conditions.

The results of this preliminary investigation areencouraging and urge us to conduct an experimentalinvestigation to verify the findings. This is currently underprogress and results will be given in the near future. Eventhough it is difficult to generalize the findings of the

current preliminary investigation, we believe that they areimportant since the reduction of soot emissions onexisting DI diesel engines is extremely important.

REFERENCES

1. Bahr,O., Karim, G.A., Liu,B., An examination of theflame spread limits in a dual fuel engine, Appl.Therm. Engng., Vol 19, pp.1071-1080, 1999

2. Karim, G.A., Khan, M.O., Examination of effectiverates of combustion heat release in a dual-fuelengine, J.S.M.E., Vol 10, No 1, 1968

3. Agarwal, A. Assanis, D.N., Multidimensionalmodeling of natural gas ignition under compressionignition conditions using detailed chemistry SAEpaper, No 980136, 1998

4. Pirouzpanah, V., Kashani, B.O., Prediction of majorpollutants emission in direct-injection dual-fuel dieseland natural-gas engines, SAE paper, No 990841,1999

5. Whitenhouse,N.D., Sareen,B.K., Prediction of heatrelease in quiescent chamber Diesel Engine allowingfor fuel/air mixing, SAE paper, No 740084, 1974

6. Kouremenos,D.A. ,Racopoulos,C.D. andHountalas,D.T. A computer simulation of combustion

process in Diesel Engines with no-swirl for thepurpose of heat release and nitric oxide prediction,Proc. Int. A.M.S.E., Vol. 3.3,207-218,1986

7. Annand, W.J.D., Heat transfer in the cylinders ofreciprocating internal combustion engines, Proc.Inst. Mech. Engrs., 177, 973-990, 1963

8. Kouremenos,D.A. ,Racopoulos,C.D. andHountalas,D.T. `Multi zone combustion modeling forthe prediction of pollutants emissions andperformance of DI Diesel engines, SAE paper, No970635, 1977.

9. Kouremenos,D.A. ,Racopoulos,C.D. andHountalas,D.T., Computer simulation with

experimental validation of the exhaust nitric oxideand soot emissions in divided chamber Dieselengines, Trans. ASME, WA meeting, San FranciscoCalifornia, Vol.10-1, pp.15-28, 1989.

10. Ramos, J.I., Internal Combustion Engine Modeling,Hemisphere, New York, 1989

11. Heywood, J.B., Internal Combustion EngineFundamentals, McGrawHill, New York, 1988

12. Benson, R.S. and Whitehouse, N.D., InternalCombustion Engines, Pergamon, Oxford, 1979

13. Hiroyasu, H., Kadota, T. and Arai, M., Developmenand use of a spray combustion modeling to predicdiesel engine efficiency and pollutant emissionsBulletin, J.S.M.E., 26, 569-576, 1983

14. Racopoulos, C.D., Hountalas, D.T., Tzanos, E.I.andTaklis, G.N.,A fast algorithm for calculating thecomposition of diesel combustion products using aneleven species chemical equilibrium schemeAdvances in Engng Software, 19, 109-119, 1994

15. Vickland, C.W., Strange, F.M., Bell, R.A. andStarkman, E.S., A consideration of the hightemperature thermodynamics of internal combustionengines, Trans. SAE, 70, 785-793, 1962GlauertM.B., The wall jet, J.Fluid Mech., 1, 625-643, 1956

16. Bazari, Z.,A DI Diesel combustion and emissionpredictive capability for use in cycle simulation, SAEpaper, No 920462, 1992

17. Glauert, M.B., The wall jet, J. Fluid Mech., 625643,1956

18. Kadota, T., Hiroyasu, H. and Oya, H., Spontaneousignition delay of a fuel droplet in high pressure andhigh temperature gaseous environments, BulletinJ.S.M.E., 19(130), 1976

19. Lavoie, G.A., Heywood, J.B. and Keck, J.C.Experimental and theoretical study of nitric oxideformation in internal combustion engines, CombustSci. and Technol.,1,313-326, 1970

20. Hountalas, D.T, Kouremenos, A.D., Developmenand application of a fully automatic troubleshootingmethod for large marine diesel engines, ApplTherm. Engng., Vol. 19, pp. 299-324, 1999

NOMENCLATURE

, b, c = constants

adel = constantA = area, m2

Af = constant in soot formation mechanism

Ab = constant in soot oxidation mechanism

Cv = specific heat capacity under constant volumeJ/kgK

Cp = specific heat capacity under constantpressure, J/kgK

CDinj = injector hole discharge coefficient

dnoz = njector hole diameter, m

D = cylinder bore, m

E = activation energy, J/Kmole

k = constant for liquid fuel preparation ratek = constant for liquid fuel reaction ratek = constant for gaseous fuel reaction ratem = mass, kg, or constant for the liquid fuel

preparation rate

= mass flow rate, kg/sec

P = pressure, Pa

PO2 = partial pressure of oxygen, bar

= volumetric flow rate, m3/sec

m!

q!


13/13

Greek

Subscripts

Dimensionless number

Abbreviations

Q = heat transfer to walls, J

R = radius, m

Rm = universal gas constant, J/kmoleK

S = penetration length, m

t = time, sec

T = absolute temperature, K

U = jet velocity, m/sec

u = specific internal energy, J/kg

V = volume, m3

X = Kmole fraction on species

y = mass fraction on species

x = constant

zi = constants

= initial jet angle, radP = pressure difference, Pa = thermal conductivity, W/mK = dynamic viscosity, kg/ms = kinematic viscosity, m2/sec = density, kg/m3 = equivalence ratio

a = air

av = available quantity

b = burned zone

del = delay

ev = evaporated liquid fuel

e = equilibrium

f = fuel

fpr = liquid fuel prepared

fi = liquid fuel injected

fupr = liquid fuel unprepared

fb = fuel burned

g = gaseous fuel or gas mixture

i = index denoting control volume or the kind ofgaseous fuel

= index denoting the species of the gaseousmixture

o = initial value

tot = total

s = soot

sf = soot formed

sb = soot oxidated

st = stoichiometric

u = unburned zone

w = wall

Re = Reynolds

DI = direct injection

bmep = brake mean effective pressure

bsfc = brake specific fuel consumption

NO = nitric oxideppm = parts per million (volume)

di dual fuel

Documents