-
Energy Conversion and Management 112 (2016) 369–381
Contents lists available at ScienceDirect
Energy Conversion and Management
journal homepage: www.elsevier .com/ locate /enconman
Energy efficiency impact of EGR on organizing clean combustionin
diesel engines
http://dx.doi.org/10.1016/j.enconman.2016.01.0420196-8904/� 2016
Elsevier Ltd. All rights reserved.
⇑ Corresponding author at: Department of Mechanical, Automotive
& MaterialsEngineering, University of Windsor, 401 Sunset
Avenue, Windsor, Ontario N9B 3P4,Canada. Tel.: +1 (519)253
3000x2636; fax: +1 (519)973 7007.
E-mail address: [email protected] (M. Zheng).
Prasad S. Divekar a, Xiang Chen b, Jimi Tjong a, Ming Zheng
a,⇑aDepartment of Mechanical, Automotive & Materials
Engineering, University of Windsor, Windsor, Ontario,
CanadabDepartment of Electrical and Computer Engineering,
University of Windsor, Windsor, Ontario, Canada
a r t i c l e i n f o a b s t r a c t
Article history:Received 30 November 2015Accepted 16 January
2016Available online 28 January 2016
Keywords:Diesel engineEGRDual-fuelEthanolUltralow NOx and
smokeThermal efficiency
Exhaust gas recirculation (EGR) is a commonly recognized primary
technique for reducing NOx emissionsin IC engines. However,
depending on the extent of its use, the application of EGR in
diesel engines isassociated with an increase in smoke emissions and
a reduction in thermal efficiency. In this work,empirical
investigations and parametric analyses are carried out to assess
the impact of EGR in attainingultra-low NOx emissions while
minimizing the smoke and efficiency penalties. Two fuelling
strategies arestudied, namely diesel-only injection and dual-fuel
injection. In the dual-fuel strategy, a high volatilityliquid fuel
is injected into the intake ports, and a diesel fuel is injected
directly into the cylinder. Theresults suggest that the reduction
in NOx can be directly correlated with the intake dilution caused
byEGR and the correlation is largely independent of the fuelling
strategy, the intake boost, and the engineload level.
Simultaneously ultra-low NOx and smoke emissions can be achieved at
high intake boost andintake dilution levels in the diesel-only
combustion strategy and at high ethanol fractions in the
dual-fuelstrategy. The efficiency penalty associated with EGR is
attributed to two primary factors; the combustionoff-phasing and
the reduction in combustion efficiency. The combustion off-phasing
can be minimized bythe closed loop control of the diesel injection
timing in both the fuelling strategies, whereas the combus-tion
efficiency can be improved by limiting the intake dilution to
moderate levels. The theoretical andempirical analyses are
summarized and the control of intake dilution and in-cylinder
excess ratio isdemonstrated for the mitigation of NOx and smoke
emissions with minimum efficiency impact.
� 2016 Elsevier Ltd. All rights reserved.
1. Introduction
The exhaust gas recirculation (EGR) is achieved by redirecting
afraction of the exhaust gases into the intake. The resulting
dilutionof the intake charge causes a significant reduction in the
engine-out NOx emissions [1–3]. Over the years, the EGR rates
appliedin diesel engines have increased consistently with the more
strin-gent NOx emission targets. Likewise, a number of EGR
configura-tions have been investigated [4,5]. Until recently, the
mostcommon EGR configuration has been the high pressure EGR
sys-tem. The high pressure EGR path connects the upstream of
thevariable geometry turbine (VGT) to downstream of the
compressor,and the VGT vanes maintain a positive backpressure to
drive theexhaust gases through the EGR path. In order to
accommodatehigher EGR flow rates without compromising the
turbocharger
efficiency, the recent trend has been to implement a dual
loopEGR configuration [6,7], where a low pressure EGR path is
addedto the existing high pressure EGR system. In the low
pressureEGR path, the treated exhaust gas from downstream of the
tur-bocharger is recirculated into the fresh air, prior to entering
thecompressor.
Although high EGR rates are common in current diesel enginesfor
NOx reduction, the smoke penalty associated with EGR remainsa
challenge (the NOx–smoke trade-off) [2,3,8]. The high
smokeemissions can be partly mitigated by advancements in the
com-mon rail fuel injection system. Research suggests that higher
injec-tion pressures and smaller nozzle diameters improve the
fuelatomization process and result in lower smoke
emissions.Increased intake boost pressure has also shown advantages
in low-ering the smoke emissions. However, the advancements in the
fuelinjection and intake boosting systems cannot completely
eliminatethe NOx and smoke trade-off [9,10]. At high engine load
levels, inparticular, the implementation of EGR for NOx control is
associatedwith an accelerated increase in the smoke, which limits
the maxi-mum allowable EGR rates at the high load conditions.
http://crossmark.crossref.org/dialog/?doi=10.1016/j.enconman.2016.01.042&domain=pdfhttp://dx.doi.org/10.1016/j.enconman.2016.01.042mailto:[email protected]://dx.doi.org/10.1016/j.enconman.2016.01.042http://www.sciencedirect.com/science/journal/01968904http://www.elsevier.com/locate/enconman
-
Nomenclature
AcronymsCA50 crank angle of 50% heat releaseCO carbon monoxideCI
compression ignitionDPF diesel particulate filterEGR exhaust gas
recirculationFSN filter smoke numberHCCI homogenous charge
compression ignitionHTC high temperature combustionIDL intake
dilution levelIMEP indicated mean effective pressureLTC low
temperature combustionMAF mass air flowNOx oxides of nitrogenSCR
selective catalytic reductionTDC top dead centerTHC total
hydrocarbons
SymbolsC1HbOc equivalent fuel molecular formulaEGRCO2 CO2 EGR
ratio
EGRmass mass EGR ratioEGRO2 O2 EGR ratiomcyl mass of cylinder
chargemegr mass of recirculated exhaust gasesnCO2 in-cylinder moles
of CO2nf in-cylinder moles of equivalent fuelnH2O in-cylinder moles
of H2OnN2 in-cylinder moles of N2nO2 in-cylinder moles of O2½O2�amb
ambient O2 concentration½O2�int cylinder O2 concentration½O2�exh
exhaust O2 concentrationR molar EGR ratiox total mole number of
combustion productsy total mole number of in-cylinder gas before
combustionz total mole number of fresh air inducted per cyclekair
fresh air excess ratiokcyl in-cylinder excess ratiow port-injection
fuel energy ratio
370 P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381
Several researchers have studied the fundamental effects of
EGRthat influence the combustion process and its impacts on the
NOxand smoke emissions. Ladommatos et al. [1] and Zheng et al.
[2]conducted experiments on diesel engines using mixtures of
purebottled gases to study the EGR composition effects. Similar
studieshave been conducted by Li et al. [11] on a natural gas spark
ignitionengine. EGR leads to a reduction in the oxygen
concentration of thecylinder charge by replacing a fraction of the
air with CO2 and H2O.The dilution of the intake charge causes the
largest reduction inNOx. However, the reduced in-cylinder oxygen is
usually associ-ated with the increase in smoke emissions.
Additionally, the higherheat capacities and the endothermic
dissociation of CO2 and H2Ocontribute to the NOx emission
reduction, albeit to a lesser extentcompared to the dilution
effect. The increased intake charge tem-perature from the mixing of
the hot exhaust gases with the freshair can cause a ‘thermal
throttling’ effect that reduces trapped gasmass, negatively impacts
the thermal efficiency and increases thesmoke emissions.
The NOx and smoke trade-off associated with the application
ofEGR has required the use of exhaust after-treatment for
meetingthe current (and future) emission standards [12]. Most
on-roaddiesel engines are equipped with a diesel particulate filter
(DPF)for smoke filtration and a selective catalytic reduction
system(SCR) for the NOx reduction. The DPF requires a periodic
injectionof fuel for regeneration of the filter and the SCR system
employsurea injection to convert the NOx into N2 and O2. The
consumptionof the supplemental fuel and the urea solution can be
reduced ifthe trade-off between the engine-out NOx and smoke
emissionscan be minimized. This will require a precise control of
the EGRamount under all operating conditions including transient
opera-tion [6]. The implementation of online EGR models [13],
analyticalapproaches [14], and high speed measurements [15] have
beeninvestigated to enable the precise EGR control.
Ultra-low levels of engine-out NOx and smoke emissions can
beattained through the implementation of low temperature
combus-tion (LTC) strategies, wherein the fuel–air mixing is
enhanced byusing early or late injection timings or by excessive
dilution. Byemploying LTC strategies, the dependence on
aftertreatment
systems can be greatly reduced. Although LTC has been
studiedextensively at steady-state conditions and moderate engine
loads[16–21], the transient and high load operation in the LTC
regimecontinues to be a challenge [22]. The air-path control
exhibits amajor hurdle in the LTC implementation. Precise control
of boostand EGR is necessary to deliver the simultaneously high
intakeboost levels and EGR rates necessary for the diesel LTC
operation[23].
The interest in renewable alternate fuels has driven theresearch
towards the combustion of bio-fuels [24,25] in dieselengines.
Studies conducted with ethanol [26], butanol [27], andvarious
biodiesels [28] have shown that EGR is necessary attainthe low NOx
emissions. The oxygen molecule contained in thesefuels lessens the
negative effect of EGR on the smoke emissions.Although the effects
of EGR have been extensively studied usingthe alternate fuels, the
impact of the fuel type on the EGR effective-ness has not been
highlighted. Under a fixed equivalence ratio, theexhaust gas
concentrations largely depend on the fuel molecularcomposition.
Hence, it is necessary to assess the impact of differentfuels on
EGR.
The interactions among the EGR rates, fuelling amounts,
fueltypes and the resultant emissions are highly complex. For a
fixedrate of EGR, e.g. 30%, the impact on the NOx and smoke
productionis altered by the engine operating conditions, as shown
in Fig. 1.The intake oxygen concentrations and NOx–smoke emissions
forthree test cases, representative of the diesel-only and the
dual-fuel combustion strategies are summarized. In case I and case
II,the diesel-only strategy is employed at low and high engine
loads,respectively. The 30% EGR is less effective in reducing the
intakeoxygen concentration, and results in high NOx emissions in
caseI. However, in case II, the same EGR rate causes a much
largerreduction in the intake oxygen concentration, which in turn
lowersthe NOx emissions and results in high smoke emissions. In
case III,the same engine load level as that of the case II is
achieved in thedual-fuel strategy by port injection of ethanol and
direct injectionof diesel. Although the same EGR rate is applied
and a similarintake oxygen concentration is achieved, the NOx and
smoke emis-sions are considerably lower in the dual-fuel case. The
port injected
-
3.7
0.09
18.2
15.6
Dual-fuel operationsupresses smoke
while EGR effectiveness in NOx reduction
increases
EGR effectiveness in NOx reduction is improved at
high load, but smoke penalty is significant
30% EGR is insufficient for effective NOx
reductionCase IIMEP: 5.2 barDiesel only
Case IIIMEP: 15.6 bar
Diesel only
Case IIIIMEP: 15.9 bar
Dual fuelDiesel+Ethanol
0.009
1.4
15.2
0.007
0.34
Smoke [g/kWh]
NOx[g/kWh]
[O2]int[%]
EGR: 30% for all test cases
Fig. 1. Illustration of the EGR impact on the intake oxygen
concentration and on theexhaust smoke and NOx emissions.
1 Compressed air supply
2 Intake surge tank3 Exhaust surge
tank4 Exhaust back-
pressure valve
5 EGR valve6 EGR cooler7 Diesel injector8 Port fuel injector9
Intake gas sampling
10 Exhaust gas sampling11 Cylinder pressure
transducer12 Optical encoder
1
6 5
2 3
8
1179 4
10
12
Portfuel
Fig. 2. Schematic of the experimental test setup.
Table 1Test engine specifications.
Model 4 Cylinder, DI Ford DuraTorq ‘‘Puma”
Displacement 1998 cm3
Bore � Stroke 86 mm � 86 mmCompression ratio 18.2:1Maximum
cylinder pressure �18 MPa (180 bar)Injection systems Common-rail
(up to 160 MPa)
P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381 371
ethanol supresses the smoke, while the NOx is also lower than
thediesel-only case.
In this work, an analytical method is presented for the study
ofEGR based on a molar balance across the air-path and a simple
in-cylinder combustion chemistry calculation. Expressions
arederived for the air–fuel ratio considering the fresh air and the
in-cylinder gases. The analysis is then extended to incorporate
theeffects of dual-fuel combustion. Thereafter, test results are
pre-sented to develop an understanding of the EGR effects on theNOx
and smoke emissions at different engine loads and differentfuelling
strategies. The impact of EGR application on thermal effi-ciency is
studied for both the diesel-only and dual-fuel modes.Based on the
EGR analysis and the experimental study, fuel strat-egy independent
EGR control considerations are laid out withlow NOx and smoke as
the primary targets and thermal efficiencyas a constraint.
2. Experimental method
The test data presented in this work is collected from a
singlecylinder, common-rail diesel engine. The engine is coupled to
aneddy current dynamometer used for load dissipation and
enginespeed control. The test platform is equipped with
independentlycontrolled intake boost, EGR, exhaust back-pressure
and fuelling(port injection and direct injection) systems. Intake
boost isachieved using an oil-free dry air compressor to generate a
highpressure combustion air supply and an electro-pneumatic valveto
regulate the boost pressure. An air flow meter measures theair
volume flow rate which is converted into a mass flow rate usingthe
measured air temperature and pressure. An intake surge tank
ismounted between the flow meter and the intake manifold to
iso-late the cyclic pulsations generated by the valve events that
wouldotherwise introduce errors in the flow rate measurement.
Exhaust
back-pressure is adjusted using a pneumatically controlled
valve.An EGR valve is used in combination with the exhaust
back-pressure valve for EGR flow regulation. An exhaust surge
tank,mounted in the exhaust path, ensures stable EGR flow and
mini-mizes the impact of exhaust pressure wave action. A
schematicof the test setup is shown in Fig. 2 and the major
specificationsof the test engine are summarized in Table 1.
A high pressure diesel common rail fuel injection system is
usedfor direct injection of diesel. The intake manifold is fitted
with asecondary port fuelling system for the port injection of the
highvolatility fuel. Two pre-calibrated gasoline port fuel
injectors sup-ply the fuel into the intake runners. During the
diesel-only tests,the common rail system is used to deliver the
diesel fuel directlyinto the combustion chamber. For the dual-fuel
tests, three highvolatility fuels, ethanol, butanol, and gasoline,
are tested for port-injection and diesel is used as the direct
injection fuel. The mainproperties of the tested fuels are listed
in Table 2. The fuel flowrates are measured using volumetric fuel
flow meters. Thesteady-state flow measurements are averaged over 60
s, whichare then converted to an average fuel mass flow rate by
applyinga fixed fuel density conversion using the fuel densities
summarizedin Table 2.
The cylinder pressure indicating system consists of a glow
plugmounted pressure transducer (AVL GU13P) and a crank
shaftmounted optical encoder with a 0.1� CA resolution. The
intakeand exhaust gas compositions are measured using a
dual-bankgas analyzer system, one for the exhaust emission
measurements(NOx, HC, CO, CO2, O2 and smoke) and the other for the
intakegas concentration measurements (CO2 and O2). The details
forthe gas analyzer system are presented in Table 3. During
everysteady state engine test point, cylinder pressure is recorded
for200 consecutive engine cycles and all other measurements
arelogged at 2 Hz and averaged over a period of 10 s.
It is noted that the results presented in this paper are subject
tosmall variations in the measurement accuracies of the
equipment
-
Table 2Test fuel specifications [29–31].
Fuel Diesel Ethanol Butanol Gasoline
Density (15 �C, kg/m3) 846 788 810 720Viscosity (30 �C, cSt) 3.5
1.52 3.5 0.64Cetane number (–) 46.5 8–11 17–25 10–17Octane number
(–) �25 110–115 87 91Lower heating value (MJ/kg) 43.5 26.9 33.1
42.4Oxygen content (% mass) 0 34.78 21.6 NegligibleBoiling Temp (1
bar, �C) 246–388 78.3 117.5 60–200Equivalent molecular formula (–)
C1H1.78 C1H3O0.5 C1H2.5O0.25 C1H1.87
Table 3Emission analyzers for emission and gas concentration
measurements.
Analyzer type Measured emissions Model
Paramagnetic O2 (%) CAI 602PHeated flame ionization THC (ppm)
CAI 300M HFIDNon-dispersive infrared CO (ppm). CO2 (%) CAI 200/300
NDIRChemiluminescence NOx (ppm) CAI 600 HCLDSmoke meter Smoke (FSN,
mg/m3) AVL Model 415S
372 P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381
used. Although the individual equipment is calibrated
regularlyand the linearity and measurement drift for the device is
typicallywithin 1%, the derived measurements presented here may
exhibitan uncertainty of up to 5%. For further details of the test
setup andthe measurements, the reader is referred to Asad et al.
[32].
3. EGR analysis
In this work, the concept of EGR is revisited from a
fundamentalunderstanding stand point. The five primary components
of thecylinder charge are considered, namely, N2, O2, CO2, H2O, and
anequivalent fuel. Concentrations of the primary components in
theintake and exhaust vary depending on the intake boost, EGR
rate,fuel type, and fuel quantity. Although the by-products of
combus-tion, such as CO, HC, NOx and smoke, are crucial from the
emissioncontrol perspective, the concentrations of these combustion
prod-ucts are usually at negligible levels for EGR ratio
calculations.Therefore, they are not included in this analysis.
3.1. Analytical approach
Using the five components of the cylinder charge, the
combus-tion reaction is written in the following form for an
equivalent fuel,C1HbOc.
nf ðC1HbOcÞ þ nO2O2 þ nN2N2 þ nH2OH2O
þ nCO2CO2��!ðnCO2 þ nf ÞCO2 þ nH2O þ b2nf� �
H2Oþ nN2N2
þ nO2 þc2nf � nf � b4nf
� �O2 ð1Þ
The intake charge comprises a mixture of N2, O2, CO2, and
H2O.The fuel is added to the mixture, which after combustion yields
theproducts that consist of the same gaseous components but in
vary-ing concentrations. The notation ‘ni’ denotes the mole number
ofthe respective specie ‘i’ before combustion. The total mole
numberof the intake charge is represented by ‘y’, while that of the
productsis denoted by ‘x’. The total number of moles of products in
Eq. (1)can be expressed in terms of the total number of moles of
the reac-tants, as shown in Eq. (2).
x ¼ yþ c2þ b4
� �nf ð2Þ
In this work, a conceptual EGR ratio is used to express the
EGRas a volumetric fraction of the total cylinder charge. Assuming
thefresh air and the circulated gas mix under isothermal
conditions,the volumetric EGR fraction is equivalent to the molar
EGR frac-tion. Thus, the molar EGR ratio (R) can be written as,
R ¼ y� zy
ð3Þ
where ‘z’ is the mole number of fresh air inducted into the
cylinder.
3.2. Quantification of EGR amount
Several methods for the evaluation of the EGR amount havebeen
implemented in academia and industry such as the one notedin Eq.
(3). The ability to quantify the EGR amount largely governsthe
development of further understanding of the EGR impact.The
frequently used EGR definitions are, therefore, summarizedin this
sub-section and their equivalence is shown.
The mass based EGR is the most commonly applied definition
intheoretical [33] and control studies [34,35]. The mass EGR
ratio(EGRmass) is the ratio of the recirculated gas mass to the
total cylin-der charge mass. The fuel mass may or may not be
included in thecylinder charge mass depending the fuel delivery
strategy. In thisstudy, the fuel mass is not considered as a part
of the gaseouscylinder contents. Nonetheless, the consideration of
fuel mass asa part of the cylinder flow introduces a minor
discrepancy in themass EGR ratio calculations. The mass EGR ratio
is expressed as,
EGRmass ¼ megrmcyl ð4Þ
where ‘megr’ is the mass of recirculated exhaust gases and
‘mcyl’ isthe cylinder charge mass. Although the mass based
definition isuseful for theoretical analysis, calculation of
EGRmass from practicalmeasurements is extremely challenging due to
the lack of accurateand robust EGR flow measurements. However, the
mass based EGRratio can be derived from the estimation of EGR flow
(orifice flowacross the EGR valve) and cylinder flow (assuming a
volumetric effi-ciency), and this approach is commonly used in
control models [36].The difference between the molar EGR ratio in
Eq. (3) and the massEGR ratio is negligible under the following
assumptions;1. The molecular weight of the intake and exhaust gases
is equal2. Molar concentration of fuel is very low (
-
P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381 373
The O2 based EGR ratio is defined as the ratio of the O2
concen-tration reduction in the intake charge to the O2
concentrationreduction in the exhaust gases, relative to the
ambient O2concentration.
EGRO2 ¼½O2�amb � ½O2�int½O2�amb � ½O2�exh
ð6Þ
In Eq. (6), ½O2�amb, ½O2�int, and ½O2�exh are the ambient,
intake, andexhaust O2 concentrations, respectively. If wet CO2 and
O2 concen-trations are considered for the EGR ratio calculations,
the EGRCO2and EGRO2 definitions are equivalent to the molar EGR
ratio (R).
3.3. Intake dilution level
Even though the EGR ratio quantifies the displacement of
thefresh air with the recycled gases, the consideration of the
effective-ness of EGR towards emission reduction is of utmost
importance.Since EGR is primarily used to reduce the NOx emissions
in dieselengines, the effectiveness of EGR can be gauged by the
extent ofNOx reduction attained by the application of EGR.
Severalresearchers have studied the individual effects of the
componentsof EGR-diluted intake charge on the NOx emissions using
bottledlaboratory gases [1–3,8]. The results suggest that the
partialreplacement of oxygen from ambient air yields the most
significantNOx reduction. Thus, an intake dilution level (IDL) is
defined hereto represent the EGR effectiveness using the O2
concentrations ofthe ambient air and the engine intake as shown in
Eq. (7). Essen-tially, the IDL is the concentration ratio between
the O2 replace-ment and the ambient O2.
IDL ¼ ½O2�amb � ½O2�int½O2�ambð7Þ
From the EGR analysis method summarized in Section 3.1, theIDL
can be correlated to the molar EGR ratio, ‘R’, by Eq. (8).
IDL ¼ R 1� ½O2�exh½O2�amb
� �ð8Þ
Since the ambient O2 concentration is nearly a constant in
mostcases, the IDL is a function of the molar EGR ratio, ‘R’, and
theexhaust oxygen concentration. The correlation between the
EGRratios and the IDL is presented in Fig. 3 for two engine
operatingconditions. The upper graph shows the calculated molar EGR
ratio,‘R’, the calculated EGRmass and the measured EGRCO2 plotted
againstthe IDL for two engine load levels in the diesel combustion
mode.
0
20
40
60
80
0
102030
4050
0 20 40 60Intake dilution level [%]
EGR
ratio
[%]
MAF EGRCO2 EGRR
Speed: 1500 rpmpint: 2 bar a
12 bar IMEP
6 bar IMEP
Speed: 1500 rpmpint: 2.5 bar aIMEP: 15 bar
Diesel+Ethanol
Diesel
EGR
ratio
[%]
Fig. 3. Relation between the intake dilution level and the EGR
ratio at differentengine load and fuel combinations.
The EGRmass is calculated from the steady-state measurement
ofthe fresh air mass flow rate and an estimation of the cylinder
flowassuming a constant volumetric efficiency. As the fuel
consump-tion increases at a higher engine load, more oxygen is
consumedduring combustion, hence reducing the oxygen in the
exhaustand in the EGR stream. As a result, the same EGR ratio
causes agreater IDL in the higher load test case.
The lower part of Fig. 3 shows two test cases under the
sameengine load and intake boost conditions but with different
fuellingstrategies. A fuel energy ratio of 1:4 (diesel:ethanol) is
used in thedual-fuel strategy. Comparing the two cases, a very
similar EGR-IDL correlation is observed regardless of the different
fuellingstrategies. The dual-fuel combustion requires a slightly
lowerEGR ratio for the same IDL due to the higher fuel quantity
requiredfor producing the same indicated load level.
3.4. Air–fuel ratio considerations
The air excess ratio defined by the actual air–fuel ratio
relativeto the stoichiometric air–fuel ratio is commonly used to
evaluatethe strength of the cylinder charge. However, this concept
needsto be revisited when the intake air is diluted with the
recirculatedexhaust gases. In order to address the effect of EGR on
the air–fuelratio, the authors have defined two excess ratio terms
in the previ-ous work [37]. The excess ratio based on the flow of
ambient air iscalled the fresh air excess ratio (kair) while the
excess ratio based onthe cylinder charge is called the in-cylinder
excess ratio (kcyl). Thein-cylinder excess ratio takes into account
the recirculation ofthe oxygen through the EGR path and hence is
greater than thefresh air excess ratio for lean combustion. Using
the EGR analysisadopted in this work, the fresh air excess ratios
can be expressedas follows,
kair ¼ 1þ Cf ½O2�exh½O2 �amb�½O2 �exh½O2 �amb
ð9Þ
Similarly, the in-cylinder excess ratio is,
kcyl ¼ 1þ Cf ½O2�exh½O2 �int�½O2 �exh½O2 �int
ð10Þ
where ‘Cf’ is a constant for a given fuel depending on the
equivalentfuel formula as shown in Eq. (11).
Cf ¼c2 þ b4� �1þ b4 � c2� � ð11Þ
Fig. 4. The correlation between the fresh air excess ratio and
the in-cylinder excessratio at varying EGR rates and engine load
levels.
-
Fig. 5. NOx reduction versus intake dilution ratio for single
injection dieselcombustion.
374 P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381
As illustrated in Fig. 4, the fresh air excess ratio and the
in-cylinder excess ratio are calculated from empirical test data of
fourEGR sweeps at different engine load levels. The excess ratios
arecalculated using the measured O2 concentrations in the intakeand
the exhaust gases (following Eqs. (9) and (10)). The color ofthe
solid dots represents the engine load level. The in-cylinderexcess
ratio is higher than the fresh air excess ratio for all testpoints
due to the recirculation of the O2 in the exhaust throughthe EGR
path. At the same engine load, the difference betweenthe
in-cylinder excess ratio and the fresh air excess ratio peaks
atintermediate EGR levels. Under low EGR conditions, the
cylindercharge primarily consists of the fresh air, and therefore
the differ-ence between the two excess ratios is insignificant. At
high EGRlevels, the combustion approaches stoichiometric conditions
andthe O2 amount in the EGR stream depletes. As a result, the
excessratio diminishes despite the high EGR flow rate. An extreme
caseis the stoichiometric burning condition caused by high EGR
usewhere no O2 is present in the EGR stream and the fresh air
excessratio is equal to the in-cylinder excess ratio.
The IDL and the excess ratio expressions are useful to
correlateengine emissions and efficiency with the effectiveness of
EGR fordifferent fuelling strategies, as discussed in later
subsections. Theexpressions for the IDL and the excess ratios
presented in Eqs.(8)–(10) can be easily applied in dual-fuel
combustion by adaptingthe fuel formulas and fuel energy ratio as
presented in Eq. (A.1) inAppendix A. However, it must be
highlighted that the equivalentmolecular formula for diesel and
gasoline fuels is solely an approx-imate representation and does
not account for the actual composi-tion of the fuels. As a result,
the molar analysis is not an accuraterepresentation of the
combustion of these multi-component fuelseven though it aids in the
understanding of the EGR impacts. Adetailed discussion of the
modification of the molar analysis tech-nique to accommodate the
dual-fuel combustion is shown inAppendix A.
4. EGR versus NOx and smoke emissions
Since the composition of the recirculated gases is dependent
onthe amount of fuel, air, and EGR, it is necessary to analyze
theimpact of EGR on the emissions under a wide range of engine
oper-ating conditions. The data presented in the following sections
isselected from a large pool of engine tests over a wide range
ofengine operating conditions and the trends are representative
ofthe entire dataset.
4.1. Single injection diesel combustion
The conventional diesel high temperature combustion (HTC)
ischaracterized by a short ignition delay with (typically) an
overlapbetween the fuel injection and the combustion events, which
pro-vides effective controllability over the combustion process
buttends to produce high NOx emissions. EGR is therefore applied
todilute the cylinder charge and achieve a significant NOx
reduction,however usually with an associated smoke penalty. A
simultane-ous lowering of the NOx and smoke emissions can be
achieved ifan adequate fuel–air mixing time is allowed by
increasing the igni-tion delay to enter into the partially premixed
LTC regimes [16,17].A long ignition delay can be attained by using
combustion phasingretard and high amounts of EGR. In order to
maintain adequatecombustion stability and to use the thermal
efficiency benefitsfrom a large expansion ratio, the phasing retard
is normally limitedto 15–25 �CA after TDC, the exact value
depending on the engineload and other operating conditions
[21,38,39]. Therefore, theenabling of simultaneously low NOx and
smoke emissions relieson heavy EGR application.
The intake dilution level through EGR is a dependent on
theengine load and the intake boost at a given EGR ratio. The
extentof NOx reduction caused by EGR is also dependent on the
engineoperation. The authors have previously shown that the
effective-ness of EGR in reducing NOx can be decoupled from other
engineoperating conditions if the intake dilution is used as a
measureof the EGR instead of the EGR ratio itself [37]. Thus, the
IDL isthe preferred parameter from the NOx control perspective.
InFig. 5, the reduction in the NOx emissions is plotted against
theIDL for test data collected over a wide range of engine
operatingconditions. The upper plot presents the measured NOx
againstthe dilution level, whereas in the lower plot the same data
set ispresented as the NOx reduction ratio normalized to the NOx
levelsat zero EGR condition for each individual case. The
representationhighlights two important trends;
a. EGR is very effective in NOx emission reduction up to an
IDLof around 25–30%. Thereafter, the effectiveness reduces
sub-stantially highlighting the challenge of achieving ultra-lowNOx
emissions.
b. Although different levels of EGR are necessary to achieve
thedesired IDL, all the test data sets tend to overlay
highlightingthe sensitivity of NOx emissions to the IDL. It must be
notedhowever, that the actual value of NOx emissions at 0%
EGRvaries with engine operating conditions, and the EGR effectis
only relative to the base NOx emission value.
While the NOx emissions depict a straightforward trend withthe
IDL, the smoke is affected by a number of engine
operatingvariables. From the air-path point of view, a major factor
thatimpacts the diesel engine smoke emissions is the cylinder
excessratio which is a function of the intake boost, EGR ratio and
the fuel-ling quantity (as shown in Eq.(10)). Since the fuelling
quantity isprimarily determined by the user torque demand, the
control overthe in-cylinder excess ratio can be exercised by a
combination ofintake boost and EGR ratio regulation.
-
Fig. 6. Effect of intake boost on smoke emissions and dilution
ratio. Fig. 7. Effect of fuelling quantity on smoke emissions and
dilution ratio.
1 For interpretation of color in Figs. 4, 5, 6, 9 and 15, the
reader is referred to theweb version of this article.
P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381 375
The smoke emissions from two EGR sweep tests are
presentedagainst the in-cylinder excess ratio in the upper plot of
Fig. 6.The fuelling amount and the fuel injection pressure are held
con-stant while the intake boost pressure is increased from 1.75
barto 2 bar abs. As shown by the test results, the lower boost
caseexhibits lower in-cylinder excess ratios throughout the EGR
sweep.When the intake pressure is 1.75 bar abs, the in-cylinder
excessratio is lower throughout the EGR sweep test. At kcyl = 1.4,
thesmoke emissions approach as high as 5 FSN which prohibits
fur-ther increase in the EGR amount. However, when the intake
boostpressure is increased to 2 bar abs, the EGR sweep curve shifts
to aleaner in-cylinder excess ratio, and the EGR can be
furtherincreased without producing excessive smoke, while avoiding
theregions of very high smoke. Following the initial increase in
theEGR amount, the reduced oxygen availability and the
loweredcombustion gas temperatures negatively impact the oxidation
rateof the smoke formed in the diesel flame, resulting in
increasedexhaust smoke emissions. However, when EGR is very high,
thediluted cylinder charge supresses the ignition of diesel such
thatthe partially premixed LTC mode is enabled and the formation
ofsmoke is avoided [21].
In the lower plot of Fig. 6, the test data is overlaid on
contours ofthe intake boost and in-cylinder excess ratio in order
to betterunderstand the impact of the intake boost level and the
EGR rateon the achievability of LTC. The contours are generated
using theEGR analysis discussed in Section 3.1. The shaded contours
repre-sent the IDL against the intake boost and the cylinder excess
ratio.Contour lines of fixed EGR ratio are also included in the
plot, and
are represented by thick blue1 lines. The reduction of NOx
requiresa sufficient IDL (�30%). However, the desired intake
dilution canreduce the in-cylinder excess ratio towards near
stoichiometriclevels when low boost is applied, e.g. in the 1.75
bar abs test case.The resultant lack of the excess in-cylinder
oxygen causes the smokeemissions to increase rapidly. When the
intake boost pressure is ele-vated, a similar intake dilution is
attained at a higher in-cylinderexcess ratio resulting in an
overall reduction in the smoke. Moreover,the higher in-cylinder
excess ratio at the same IDL allows furtherEGR increase which
ultimately enables the LTC operation.
The impact of increasing the engine load (fuelling amount) onthe
exhaust smoke emissions is presented in Fig. 7. The smokeemissions
are plotted against the cylinder excess ratio at threeengine load
levels in the upper part of Fig. 7 for fixed intake boostand fuel
injection pressures. At the low load condition (i.e. the6 bar IMEP
test case), the cylinder charge is over lean and as theEGR is
increased, the in-cylinder excess ratio decreases causingthe smoke
to increase. The peak smoke, however, is relativelylow as a result
of the sufficiently large amount of in-cylinder oxy-gen. With
further EGR application, LTC is enabled during which thein-cylinder
excess ratio remains sufficiently lean (kcyl � 2). Whenthe fuelling
amount is increased to achieve 10 bar IMEP, the EGRsweep curve
shifts towards a richer air–fuel ratio operation.Consequently, the
smoke emissions increase more rapidly and alarger smoke peak is
observed compared to the low load test case.
-
376 P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381
Nevertheless, the in-cylinder excess ratio is sufficiently lean
at thesmoke peak such that additional EGR rates can be applied to
enableLTC where sharp drop of smoke is observed. When the engine
loadis further increased to 12 bar IMEP, the in-cylinder excess
ratioapproaches 1.0 as the EGR ratio is increased. The lack of
oxygen(low in-cylinder excess ratio) at the smoke peak prohibits
furtherEGR application. It must be noted however, that other
researchershave achieved LTC by operating in the fuel-rich region
(smokelessrich combustion) [40]. Fuel rich operation is not
attempted in thecurrent work in the interest of maintaining
acceptable combustionefficiency and limiting the maximum smoke peak
during thetransition.
In the lower plot of Fig. 7, the EGR sweep test data at the
threeengine load levels is marked onto the IDL and EGR ratio
contoursgenerated from parametric calculations. The IDL is
presented inthe form of colored contours against the fuelling
amount and thein-cylinder excess ratio. Lines of fixed EGR ratio
are also overlaidon the chart. At lower fuelling rates, the
in-cylinder excess ratiois higher and thus a larger intake dilution
can be attained byincreasing the EGR amount. However, as the
fuelling amountincreases, the window of lean operation shrinks
rapidly, thusrequiring a lower air–fuel ratio to attain the same
IDL. It may alsobe noted that at the lower fuelling rates, a
significantly higher EGRratio is necessary to achieve a certain
level of intake dilutionwhereas when the fuelling rate increases,
the same IDL can beattained at a lower EGR ratio.
In Fig. 8, the indicated NOx and smoke emissions are plotted
forthree EGR sweeps to demonstrate the NOx–smoke trade-off andthe
enabling of LTC in the single intake diesel combustion strategy.To
summarize the NOx–smoke trade-off and the enabling of LTC inthe
single injection diesel combustion strategy, the indicated NOxand
smoke emissions are plotted in Fig. 8 for three EGR sweeptests. The
EGR sweeps are conducted at three engine load levelswherein the
intake boost and fuel injection pressure are concur-rently
increased at the higher engine load. The emissions data isplotted
on a log scale for easy readability. Dotted lines of constantIDL
are marked onto the data to highlight the test points with sim-ilar
IDL. As highlighted in Fig. 6, the indicated NOx emissions
arelargely insensitive to the load level but primarily depend on
theIDL. The smoke emissions exhibit an increasing trend as the
engineload increases, even though the intake boost and fuel
injectionspressures are raised. In the 5 bar and 10 bar IMEP test
cases, theIDL can be increased to enable LTC, whereas in the 16 bar
test case,further increase in EGR is restricted excessive smoke
emissions.(6 FSN at 28% IDL)
Fig. 8. NOx–smoke trade-off and LTC enabling for single
injection dieselcombustion.
4.2. Diesel injection with port fuelling
Diesel fuel’s low volatility and its strong tendency for
auto-ignition (high cetane number) limit the ability to enable
partiallypre-mixed combustion at increased load levels. The
dual-fuel strat-egy has been identified as a promising solution to
extend theengine load level in the low NOx and smoke combustion
regime.This strategy includes the injection of a high volatility,
low cetanefuel at the intake port followed by the in-cylinder high
pressureinjection of diesel fuel. Under the dual-fuel combustion
strategy,the application of EGR is necessary for NOx emission
reductionwhereas the pre-mixed portion of the cylinder charge may
reducethe smoke penalty associated with the increase in the EGR
amount.
EGR sweep tests conducted with three high volatility fuels inthe
dual-fuel mode are presented in Fig. 9. The high volatility
fuel(ethanol, butanol, or gasoline) is port-injected during the
inductionstroke followed by the near top dead center (TDC)
injection of die-sel. The ratio of the two fuels is adjusted such
that 45–50% of thetotal fuel energy is contributed by the port
injected fuel. Testresults for the diesel-only combustion
configuration at the sameoperating conditions are also plotted for
reference. Compared tothe diesel-only results, when diesel is
partially replaced with theport injected fuel, NOx emissions
decrease, even at 0% IDL.The impact of the dual-fuel strategy on
NOx emissions canbe explained from two main aspects. A larger
fraction of thecombustion is lean and pre-mixed in the dual-fuel
mode causinga reduction in NOx emissions. In addition, the
evaporation ofthe port-injected fuel during the compression stroke
causes anoticeable reduction in the compression-end temperature,
henceresulting in a lower flame temperature. The effect of the
dual-fuel strategy on NOx emissions is the largest in the
gasoline–dieseltest case among the cases presented in Fig. 9.
Analysis of thecylinder pressure data reveals that the combustion
duration inthe gasoline–diesel test case is the longest compared to
the otherdual-fuel test cases. The long combustion duration
contributes toa lower temperature rise which may result in the
lower NOx.
As shown by the overall trend of all four presented cases,
NOxemissions drop substantially as the IDL is increased at
higherEGR rates. Therefore, the application of EGR is still the
primaryenabler of ultra-low NOx emissions in either diesel-only or
dual-fuel combustion. When the NOx emission reduction is
presentedrelative to the peak NOx emissions at 0% IDL, all the test
cases fol-low a similar trend. This highlights the correlation
between intake
Fig. 9. NOx reduction versus intake dilution level for dual-fuel
combustion.
-
Fig. 10. Smoke emission trends for dual-fuel combustion.
Fig. 11. Heat release traces for selected test points in the
dual-fuel combustionmode.
Fig. 12. NOx–smoke trade-off and LTC enabling in diesel–ethanol,
dual-fuelcombustion.
P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381 377
dilution and NOx reduction, which can be applied to both
thediesel-only and dual-fuel strategies.
The smoke emissions, however, do not exhibit a consistenttrend
when different fuels are used in dual-fuel combustion.
Whenapproximately 50% of the diesel fuel is replaced by the port
injec-tion of ethanol, a noticeable reduction in the smoke
emissions isobserved throughout the EGR sweep test. This effect
however isnot seen when butanol or gasoline is used at a similar
fuel ratio.In fact, with the application of EGR, the smoke
emissions tend toincrease more sharply in the diesel–gasoline and
diesel–butanoltesting cases (see Fig. 10).
The heat release rate traces for a representative case of each
ofthe three dual-fuel combustion tests are shown in Fig. 11
toexplain the trends in the smoke emissions. Ethanol has a
strongtendency to resist auto-ignition even on the high compression
ratioengine due to its high octane number. Thus, the combustion of
thepremixed ethanol-air charge is essentially initiated after the
dieselfuel delivery has commenced. On the other hand, when butanol
orgasoline is port injected, the low auto-ignition resistance of
thefuel (relative to that of ethanol) causes a portion of the
portinjected fuel to auto ignite prior to the diesel injection.
Thus, the
ignition delay of the diesel is substantially shortened and the
dieselundergoes diffusion type combustion, resulting in a
significant risein the smoke emissions.
These test results suggest that with the high compression
ratio,the dual-fuel combustion strategy using a relatively low
octane(e.g. butanol or gasoline) still encounters the NOx–smoke
trade-off. However, the auto-ignition of port injected fuels
indicates thepossibility of enabling HCCI type combustion, which is
beyondthe scope of this work and more details can be found in
theauthors’ previous work [41,42].
Nevertheless, the test results of ethanol–diesel combustionhave
clearly shown the benefits of implementing a dual-fuel strat-egy in
addition to the EGR application for the reduction of smokeand NOx
emissions. The results of the ethanol–diesel combustionat higher
ethanol fractions are summarized in Fig. 12. Under thesame engine
boundary conditions, EGR sweeps are performed atfour different fuel
ratios. The NOx and smoke emissions are plottedon the log scale.
The ethanol substitution not only supresses thetendency to produce
smoke at elevated EGR rates, but the NOxemissions are also reduced
as the ethanol fraction is increased sug-gesting that a lower EGR
may be sufficient for achieving the sametarget NOx emissions. At
very high ethanol fractions, e.g. 80%, thesmoke emissions remain
ultra-low, and the increase in the EGRrate produces a consistent
reduction in the NOx emissions.
5. EGR and thermal efficiency
Previous research has suggested that the use of EGR bears
anegative impact on the specific fuel consumption in diesel
engines[4,43,44]. The reduction in the thermal efficiency is
attributed toseveral factors that can be broadly categorized as the
combustionand system level factors. The pumping work associated
with a pos-itive exhaust-to-intake manifold pressure difference
(required todrive sufficient EGR across the EGR path) is the
primary contribu-tor to the system level reduction in the
efficiency. The combustionrelated factors, on the other hand,
comprise of the reduced cyclework and deteriorated combustion
efficiency. The reduction inthe cycle work is primarily caused by
the retarded combustionphasing with EGR application if the diesel
injection timing is fixed,whereas, the reduction in the combustion
efficiency is caused bythe increase in the exhaust HC and CO
emissions. Other factorssuch as prolonged combustion duration,
lowered air-excess ratioand reduced combustion temperature also
contribute to the effi-ciency loss. This work mainly focuses on the
engine efficiency from
-
Fig. 13. Effect of EGR on combustion phasing control by diesel
injection command.
378 P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381
the combustion perspective as discussed in the
subsequentsubsections.
Fig. 14. Effect of EGR on thermal efficiency for diesel
combustion and diesel–ethanol dual-fuel combustion.
5.1. Combustion phasing impact on efficiency
The ignition delay is typically prolonged as EGR is
increasinglyapplied. When the injection timing is fixed, the start
of combustionpostpones in accordance to the longer ignition delay
and the entirecombustion event can be delayed into the expansion
stroke. Enginetest results are presented to demonstrate the
combustion phasingretard caused by EGR application and its effect
on the engine ther-mal efficiency. The results of two EGR sweeps
are presented inFig. 13. These two test sets are conducted under
the same engineconditions except the fuel injection timing. The
CA50 and thermalefficiency are plotted against the NOx emissions so
that the impactin achieving the NOx reduction is highlighted. The
solid dots repre-sent the results where the diesel injection timing
was held con-stant throughout the EGR sweep test, while the hollow
dotspresent test results wherein the fuel injection timing was
adjustedto maintain the CA50 within 1 �CA of the baseline CA50.
As the EGR rate is increased while the fuel injection timing
isfixed, the ignition delay increases and the CA50 shifts later.
Whenthe same EGR sweep test is repeated with a fixed CA50, majority
ofthe thermal efficiency penalty is recovered. It is noted that a
grad-ual reduction in the thermal efficiency is observed as EGR
isincreased, even when the CA50 is fixed. The thermal
efficiencyreduction is more prominent at high IDL applied to
achieve NOxemissions lower than 0.2 g/kW h. This reduction in the
thermalefficiency may be caused by the reduction in combustion
effi-ciency, increase in combustion duration, a lower
combustiontemperature.
5.2. Combustion efficiency impact
In the diesel-only combustion, the exhaust HC and CO emis-sions
tend to increase sharply when heavy EGR is applied, whereasin the
dual-fuel combustion these incomplete combustion prod-ucts remain
at high levels across the EGR range. The increasedHC and CO
emissions lead to lower combustion efficiency and inturn reduce the
engine thermal efficiency. The EGR and dual-fuelstrategy impacts on
the combustion efficiency are investigated inthe current section.
Representative results from two EGR sweeptests are presented in
Fig. 14. The diesel-only fuelling strategyand the ethanol–diesel
strategy are tested at 10 bar IMEP. In boththe test cases, the CA50
is maintained at 368–369 �CA by adjustingthe diesel injection
timing to counteract the combustion phasingshift from the use of
EGR. The smoke emissions, combustion effi-ciency and thermal
efficiency are plotted against NOx emissions.
In the diesel-only case, the initial reduction in NOx (from
highlevel to 0.3 g/kW h) by increasing the EGR rate does not incur
anoticeable penalty in the smoke or the combustion
efficiency,although the thermal efficiency shows a minor decrease,
consistentwith trends reported in Section 5.1. This decrease may
beattributed to the marginal increase in combustion duration andthe
reduction in the combustion temperature. A further increasein EGR
reveals the NOx–smoke trade-off and the thermal efficiencycontinues
to decrease. A gradual reduction in the combustionefficiency also
contributes to the thermal efficiency reduction atthis point. As
the intake dilution is increased further to attainsimultaneous
ultra-low NOx and smoke, the thermal efficiencypenalty increases
substantially because of a sharp rise in theincomplete combustion
products. It is noted that ultra-low NOx(>0.2 g/kW h) can be
achieved without a significant thermal effi-ciency penalty, but
further EGR addition is necessary to attainultra-low smoke
emissions in LTC where significant efficiencydegradation is
observed.
Under the ethanol–diesel combustion, the NOx reduction byEGR can
be obtained with a smaller rise in smoke. When a highethanol
fraction is used, ultra-low smoke emissions can beachieved as a
larger portion of the cylinder charge is already pre-mixed (recall
Fig. 12). The high homogeneity also results inreduced combustion
efficiency, as seen in Fig. 14. However, inthe dual-fuel mode, the
thermal efficiency is marginally higherthan that of the diesel-only
mode. The combustion inefficiency iscompensated by the thermal
efficiency gain from the charge cool-ing effect of the port
injected ethanol [45]. It may be noted thatalthough EGR is
necessary for NOx abatement in both the combus-tion strategies
presented in Fig. 14, the low smoke in the dual-fuelmode reduces
the IDL necessary for achieving LTC (simultaneouslylow NOx and
smoke). Thus the rapid reduction in thermal effi-ciency at LTC
conditions is not observed in the dual-fuel strategy.
6. EGR control considerations
The engine test results presented in this work indicate that
inthe diesel-only mode the NOx–smoke emission trade-off remainsa
major challenge in deciding the applicable EGR amount. LTCmay be
enabled with excessively high rates of EGR. This requires
-
Fig. 15. Contour map of intake boost pressure and MAF at varying
engine loads anda fixed IDL.
Fig. 16. Engine full load demonstrated with diesel ignited
ethanol LTC.
P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381 379
significantly larger intake boost amounts to maintain
sufficientlyhigh in-cylinder excess ratio while attaining the
desired intakedilution. Furthermore, a large thermal efficiency
penalty isobserved when excess amounts of EGR are applied to attain
theLTC conditions. On the contrary, in the dual-fuel mode the
dis-placement of diesel with the port injected ethanol supresses
thesmoke emissions when a large ethanol fraction is used. Even
inthe dual-fuel mode, sufficient intake dilution by EGR is
necessaryfor NOx reduction and a high in-cylinder air-excess ratio
isdesirable.
While the EGR flow displaces the fresh intake air, maintaining
alean in-cylinder charge requires adequate control over the
intakeboost and the EGR flow rate. As the EGR flow rate is
typically notmeasured, it is necessary to control the intake boost
and the freshmass air flow (MAF) rate. In this regard, a contour
map of intakeboost and MAF is presented in Fig. 15. The map is
generated fromthe parametric EGR analysis discussed in Section 3.
The analysis isconducted at a fixed IDL (for the most effective NOx
reduction,recall Figs. 5 and 9) over a range of boost, MAF and fuel
amounts.The color contours represent the IMEP levels corresponding
tothe increased fuelling amounts. Iso-lines of EGR (solid blue)
andin-cylinder excess ratio (dotted orange) are overlaid on the
samecontour plot.
The trends presented in Fig. 15 can be understood as
follows.While maintaining a fixed IDL at a low engine load level
(bottomright region), a high cylinder excess ratio can be attained
byincreasing the intake boost level. However, the EGR rate
necessaryfor this condition rapidly increases which may be achieved
byincreasing the intake boost pressure and maintaining a fixedMAF.
On the contrary, at the high load condition (top right region),the
same IDL is achieved at a much lower EGR ratio. The intakeboost
pressure and the MAF have to be increased simultaneouslyas the load
level increases. If a fixed IDL is desired, in combinationwith a
fixed in-cylinder excess ratio, a linear relationship isobserved
between the intake boost level, fresh air mass flow rateand the
fuelling amount. The boost, MAF and fuelling correlationis similar
to the EGR and air excess ratio correlation presented byNakayama et
al. [46]. While the EGR quantities necessary to main-tain a fixed
dilution and an in-cylinder excess ratio may vary withthe engine
load, a closed loop control over intake boost and thefresh air flow
may be designed to achieve the NOx reduction with-out compromising
the in-cylinder excess ratio.
A full load test point operated in the diesel–ethanol mode
isoverlaid on the contour map to further explain the utility of
such
a representation. The cylinder pressure and heat release trace
forthis test point are presented in Fig. 16. The full load
condition atthe desired IDL is achieved by operating
close-to-stoichiometricconditions at 2.5 bar abs intake pressure.
At this test condition, fur-ther increase in the intake boost is
prevented by the peak cylinderpressure limit of the test
engine.
The operating window for the fixed IDL shrinks as the engineload
increases. Moreover, significantly higher levels of intake
boostpressure may be necessary to maintain the dilution and
in-cylinderexcess ratios at high engine load points. In reality,
the ability toincrease the intake boost is limited by the
turbocharging hardwareas well as the peak cylinder pressure limit
of the engine. Neverthe-less, this EGR analysis provides guidelines
for better operating theEGR and turbocharging systems when
pre-defined limits for intakedilution and in-cylinder excess ratio
are available.
7. Conclusions
Empirical studies are conducted to develop a
comprehensiveunderstanding of the use of EGR for achieving
ultra-low NOx emis-sions. The impact of the EGR rate on the engine
performanceparameters is investigated for two fuelling strategies,
the diesel-only strategy and the dual-fuel strategy. An EGR
analysis is devel-oped to account for the different fuel molecular
structures anddual-fuel applications. The research results are
summarized asfollows:
1. Using the EGR analysis and test data, an engine
operatingparameter IDL is defined as an indicator to gauge the
effective-ness of EGR on NOx reduction independent of engine
operatingconditions and fuelling strategies.
2. In the dual-fuel combustion mode, three port-injection fuels
areinvestigated, ethanol, butanol, and gasoline, while a
moderateamount of diesel e.g. 12–50% is used as the
direct-injection fuel.Although lower NOx emissions are observed
when a larger frac-tion of the port-injected fuel is used, EGR is
still necessary toachieve the ultra-low NOx emissions.
3. The smoke emissions tend to increase as the intake
dilutionlevel increases for both diesel-only and dual-fuel
combustionstrategies. However, in the dual-fuel combustion, the
ethanol–diesel combustion results in a reduction of this smoke
penalty.
4. In addition to the conventional air excess ratio, the
in-cylinderexcess ratio is used to explain the trends of the smoke
emis-sions associated with the increase in the EGR. A larger
in-cylinder excess ratio is beneficial for the lowering of the
smokeemissions and for achieving LTC in the diesel-only
combustionmode.
-
380 P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381
5. The reduction in the thermal efficiency associated with
theincrease in the EGR rate can be attributed to the
combustionoff-phasing and the combustion efficiency
degradation.
6. The combustion-off phasing caused by EGR increase can
becompensated by adjusting the diesel injection timing in boththe
diesel-only and the diesel–ethanol dual-fuel strategies.
7. At 10 bar IMEP, a heavy use of EGR is necessary to attain
simul-taneously low smoke and NOx emissions in the
diesel-onlycombustion mode, at which point the combustion
efficiencydecreases sharply. In the diesel–ethanol dual-fuel
combustionmode, heavy EGR application is not necessary as the
smokeemissions are ultra-low when a high ethanol fraction is
used.By avoiding the heavy use of EGR, the large reduction in
com-bustion efficiency is avoided.
8. The intake dilution and in-cylinder excess ratio control
byintake boost and fresh air flow regulation is suggested for
theattainment of the desired NOx reduction with a smaller
smokepenalty. The quantification of EGR in this can facilitate to
over-come the challenges associated with the EGR flow
adjustment.
Acknowledgements
The research is supported by NSERC CRD, Discovery,
CREATEprograms; the NCE AUTO21 and BioFuelNet programs; the
FordMotor Company, and the University of Windsor.
Appendix A. EGR analysis for dual-fuel mode
When the dual-fuel combustion mode is considered, EGR anal-ysis
can be complicated due to the different fuel compositions andthe
varying fuel quantities. The EGR analysis for the dual-fuel
sce-nario can be greatly simplified by defining an equivalent
hypothet-ical fuel C1HbOc. The hypothetical fuel produces the same
molenumbers of the primary exhaust gas components when it
replacesthe two test fuels.
The two fuels used for the current dual-fuel analysis are
repre-sented as follows.
1. C1Hb1Oc12. C1Hb2Oc2
In the dual-fuel mode, an energy ratio, w, is used to evaluate
therelative amounts of the two fuels used.
w ¼ nf2LHV2nf1LHV1 þ nf2LHV2 ðA:1Þ
where LHV1 and LHV2 (represented in energy per moles of fuel)
arethe lower heating values of the primary and secondary fuels.
Amolar fuel ratio, w0, is defined to evaluate the fuel mole
numberratio of the secondary fuel to the total moles of fuel.
w0 ¼ nf2nf1 þ nf2 ðA:2Þ
Combustion for the dual-fuel mode can be written as,
nf1ðC1Hb1Oc1Þ þ nf2ðC1Hb2Oc2Þ þ nO2O2 þ nN2N2 þ nH2OH2Oþ nCO2CO2
! ðnCO2 þ nf1 þ nf2ÞCO2þ nH2O þ
b12
nf1 þ b22 nf2� �
H2Oþ nN2N2
þ nO2 þc12
nf1 þ c22 nf2 � nf1 � nf2 �b14
nf1 � b24 nf2� �
O2 ðA:3Þ
By comparing Eq. (A.3) to Eq. (1), the following can be
derived
nf ¼ nf1 þ nf2 ðA:4Þ
b ¼ b2 � w0 þ b1ð1� w0Þ ðA:5Þ
c ¼ c2 � w0 þ c1ð1� w0Þ ðA:6ÞUsing the equivalent fuel
formulation, if the fuel ratio is known,
the EGR analysis approach developed in Section 3 including the
airexcess ratio definitions apply to the dual-fuel mode.
References
[1] Ladommatos N, Abdelhalim S, Zhao H. The effects of exhaust
gas recirculationon diesel combustion and emissions. Int J Engine
Res
2000;1(1):107–26.http://dx.doi.org/10.1243/1468087001545290.
[2] Zheng M, Reader GT, Hawley JG. Diesel engine exhaust gas
recirculation––areview on advanced and novel concepts. Energy
Convers Manage 2004;45(6):883–900.
http://dx.doi.org/10.1016/S0196-8904(03)00194-8.
[3] Maiboom A, Tauzia X, Hétet J-F. Experimental study of
various effects ofexhaust gas recirculation (EGR) on combustion and
emissions of anautomotive direct injection diesel engine. Energy
2008;33(1):22–34.
http://dx.doi.org/10.1016/j.energy.2007.08.010.
[4] Baert RS, Beckman DE, Veen A. Efficient EGR technology for
future HD dieselengine emission targets. SAE technical paper
1999-01-0837; 1999. http://dx.doi.org/10.4271/1999-01-0837.
[5] Millo F, Giacominetto PF, Bernardi MG. Analysis of different
exhaust gasrecirculation architectures for passenger car diesel
engines. Appl Energy2012;98:79–91.
http://dx.doi.org/10.1016/j.apenergy.2012.02.081.
[6] Shutty J, Czarnowski R. Control strategy for a dual loop EGR
system to meetEuro 6 and beyond. In: Dir engine-effic emiss reduct
res DEER conf; 2009.
[7] Huang Y, Colvin J, Wijesinghe A, Wang M, Hou D, Fang Z. Dual
loop EGR inretrofitted heavy-duty diesel application. SAE technical
paper 2014-01-1244;2014.
http://dx.doi.org/10.4271/2014-01-1244.
[8] Abd-Alla GH. Using exhaust gas recirculation in internal
combustion engines: areview. Energy Convers Manage
2002;43(8):1027–42.
http://dx.doi.org/10.1016/S0196-8904(01)00091-7.
[9] Aoyagi Y, Kunishima E, Asaumi Y, Aihara Y, Odaka M, Goto Y.
Dieselcombustion and emission using high boost and high injection
pressure in asingle cylinder engine (effects of boost pressure and
timing retardation onthermal efficiency and exhaust emissions).
JSME Int J Ser B 2005;48(4):648–55.
http://dx.doi.org/10.1299/jsmeb.48.648.
[10] Natti K, Sinha A, Hoerter C, Andersson P, Andersson J,
Lohmann C, et al. Studieson the impact of 300 MPa injection
pressure on engine performance, gaseousand particulate emissions.
SAE Int J Engines 2013;6:336–51.
http://dx.doi.org/10.4271/2013-01-0897.
[11] Li W, Liu Z, Wang Z, Xu Y. Experimental investigation of
the thermal anddiluent effects of EGR components on combustion and
NOx emissions of aturbocharged natural gas SI engine. Energy
Convers Manage2014;88:1041–50.
http://dx.doi.org/10.1016/j.enconman.2014.09.051.
[12] Johnson TV. Review of diesel emissions and control. SAE Int
J Fuels Lubr 2010;3(1):16–29.
http://dx.doi.org/10.4271/2010-01-0301.
[13] Zeng X, Wang J. A physics-based time-varying transport
delay oxygenconcentration model for dual-loop exhaust gas
recirculation (EGR) engineair-paths. Appl Energy 2014;125:300–7.
http://dx.doi.org/10.1016/j.apenergy.2014.03.076.
[14] Asad U, Tjong J, Zheng M. Exhaust gas recirculation – zero
dimensionalmodelling and characterization for transient diesel
combustion control.Energy Convers Manage 2014;86:309–24.
http://dx.doi.org/10.1016/j.enconman.2014.05.035.
[15] Regitz S, Collings N. Fast response air-to-fuel ratio
measurements using anovel device based on a wide band lambda
sensor. Meas Sci Technol2008;19:075201.
http://dx.doi.org/10.1088/0957-0233/19/7/075201.
[16] Ogawa H, Li T, Miyamoto N. Characteristics of low
temperature and lowoxygen diesel combustion with ultra-high exhaust
gas recirculation. Int JEngine Res 2007;8(4):365–78.
http://dx.doi.org/10.1243/14680874JER00607.
[17] Kimura S, Aoki O, Ogawa H, Muranaka S, Enomoto Y. New
combustion conceptfor ultra-clean and high-efficiency small DI
diesel engines. SAE technical paper1999-01-3681; 1999.
[18] Hasegawa R, Yanagihara H. HCCI combustion in DI diesel
engine. SAE technicalpaper 2003-01-0745. Warrendale, PA; 2003.
[19] Jacobs TJ, Assanis DN. The attainment of premixed
compression ignition low-temperature combustion in a compression
ignition direct injection engine.Proc Combust Inst
2007;31(2):2913–20.
http://dx.doi.org/10.1016/j.proci.2006.08.113.
[20] Zheng M, Han X, Reader GT. Empirical studies of EGR enabled
diesel lowtemperature combustion. J Automot Saf Energy
2010;1(3):219–28.
[21] Dec JE. Advanced compression-ignition engines—understanding
the in-cylinder processes. Proc Combust Inst 2009;32(2):2727–42.
http://dx.doi.org/10.1016/j.proci.2008.08.008.
[22] Kodama Y, Nishizawa I, Sugihara T, Sato N, Iijima T,
Yoshida T. Full-load HCCIoperation with variable valve actuation
system in a heavy-duty diesel engine.SAE technical paper
2007-01-0215; 2007.
[23] Asad U, Zheng M. Tightened intake oxygen control for
improving diesel low-temperature combustion. Proc Inst Mech Eng
Part J Automob Eng2011;225:513–30.
http://dx.doi.org/10.1177/2041299110393211.
http://dx.doi.org/10.1243/1468087001545290http://dx.doi.org/10.1016/S0196-8904(03)00194-8http://dx.doi.org/10.1016/j.energy.2007.08.010http://dx.doi.org/10.1016/j.energy.2007.08.010http://dx.doi.org/10.4271/1999-01-0837http://dx.doi.org/10.4271/1999-01-0837http://dx.doi.org/10.1016/j.apenergy.2012.02.081http://dx.doi.org/10.4271/2014-01-1244http://dx.doi.org/10.1016/S0196-8904(01)00091-7http://dx.doi.org/10.1016/S0196-8904(01)00091-7http://dx.doi.org/10.1299/jsmeb.48.648http://dx.doi.org/10.4271/2013-01-0897http://dx.doi.org/10.4271/2013-01-0897http://dx.doi.org/10.1016/j.enconman.2014.09.051http://dx.doi.org/10.4271/2010-01-0301http://dx.doi.org/10.1016/j.apenergy.2014.03.076http://dx.doi.org/10.1016/j.apenergy.2014.03.076http://dx.doi.org/10.1016/j.enconman.2014.05.035http://dx.doi.org/10.1016/j.enconman.2014.05.035http://dx.doi.org/10.1088/0957-0233/19/7/075201http://dx.doi.org/10.1243/14680874JER00607http://dx.doi.org/10.1016/j.proci.2006.08.113http://dx.doi.org/10.1016/j.proci.2006.08.113http://refhub.elsevier.com/S0196-8904(16)00059-5/h0100http://refhub.elsevier.com/S0196-8904(16)00059-5/h0100http://dx.doi.org/10.1016/j.proci.2008.08.008http://dx.doi.org/10.1016/j.proci.2008.08.008http://dx.doi.org/10.1177/2041299110393211
-
P.S. Divekar et al. / Energy Conversion and Management 112
(2016) 369–381 381
[24] Agarwal AK. Biofuels (alcohols and biodiesel) applications
as fuels for internalcombustion engines. Prog Energy Combust Sci
2007;33:233–71. http://dx.doi.org/10.1016/j.pecs.2006.08.003.
[25] Han X, Zheng M, Wang J. Fuel suitability for low
temperature combustion incompression ignition engines. Fuel
2013;109:336–49. http://dx.doi.org/10.1016/j.fuel.2013.01.049.
[26] Padala S, Woo C, Kook S, Hawkes ER. Ethanol utilisation in
a diesel engineusing dual-fuelling technology. Fuel
2013;109:597–607. http://dx.doi.org/10.1016/j.fuel.2013.03.049.
[27] Chen Z, Liu J, Wu Z, Lee C. Effects of port fuel injection
(PFI) of n-butanol andEGR on combustion and emissions of a direct
injection diesel engine. EnergyConvers Manage 2013;76:725–31.
http://dx.doi.org/10.1016/j.enconman.2013.08.030.
[28] Torregrosa AJ, Broatch A, Plá B, Mónico LF. Impact of
Fischer–Tropsch andbiodiesel fuels on trade-offs between pollutant
emissions and combustionnoise in diesel engines. Biomass Bioenergy
2013;52:22–33.
http://dx.doi.org/10.1016/j.biombioe.2013.03.004.
[29] Alternative Fuels Data Center – Fuel Properties Comparison;
n.d. [accessed January 6, 2016].
[30] Dillon HE, Penoncello SG. A fundamental equation for
calculation of thethermodynamic properties of ethanol. Int J
Thermophys
2004;25:321–35.http://dx.doi.org/10.1023/B:IJOT.0000028470.49774.14.
[31] Szwaja S, Naber JD. Combustion of n-butanol in a
spark-ignition IC engine. Fuel2010;89:1573–82.
http://dx.doi.org/10.1016/j.fuel.2009.08.043.
[32] Asad U, Kumar R, Han X, Zheng M. Precise instrumentation of
a diesel single-cylinder research engine. Measurement
2011;44:1261–78.
http://dx.doi.org/10.1016/j.measurement.2011.03.028.
[33] Shyani RG, Caton JA. A thermodynamic analysis of the use of
exhaust gasrecirculation in spark ignition engines including the
second law ofthermodynamics. Proc Inst Mech Eng Part J Automob Eng
2009;223:131–49.http://dx.doi.org/10.1243/09544070JAUTO935.
[34] Kolmanovsky I, Morall P, Van Nieuwstadt M, Stefanopoulou A.
Issues inmodelling and control of intake flow in variable geometry
turbochargedengines. Chapman Hall CRC Res Notes Math
1999:436–45.
[35] Van Nieuwstadt MJ, Kolmanovsky IV, Moraal PE, Stefanopoulou
A, Jankovic M.EGR-VGT control schemes: experimental comparison for
a high-speed dieselengine. Control Syst IEEE 2000;20:63–79.
[36] Guzzella L, Onder CH. Introduction to modeling and control
of internalcombustion engine systems. Berlin, Heidelberg: Springer;
2010.
[37] Asad U, Zheng M. Exhaust gas recirculation for advanced
diesel combustioncycles. Appl Energy 2014;123:242–52.
http://dx.doi.org/10.1016/j.apenergy.2014.02.073.
[38] Kook S, Bae C, Miles PC, Choi D, Pickett LM. The influence
of charge dilutionand injection timing on low-temperature diesel
combustion and emissions.SAE technical paper 2005-01-3837.
Warrendale, PA; 2005.
[39] d’Ambrosio S, Ferrari A. Effects of exhaust gas
recirculation in diesel enginesfeaturing late PCCI type combustion
strategies. Energy Convers Manage2015;105:1269–80.
http://dx.doi.org/10.1016/j.enconman.2015.08.001.
[40] Akihama K, Takatori Y, Inagaki K, Sasaki S, Dean AM.
Mechanism of thesmokeless rich diesel combustion by reducing
temperature. Warrendale,PA: SAE International; 2001.
[41] Han X, Wang M, Zheng M. An enabling study of neat n-butanol
HCCIcombustion on a high compression-ratio diesel engine. SAE
technical paper2015-01-0001. Warrendale, PA; 2015.
[42] Han X, Xie K, Zheng M, De Ojeda W. Ignition control of
gasoline–diesel dualfuel combustion. SAE technical paper
2012-01-1972. Warrendale, PA; 2012.
[43] Kohketsu S, Mori K, Sakai K, Hakozaki T. EGR technologies
for a turbochargedand intercooled heavy-duty diesel engine. SAE
technical paper 970340; 1997.
[44] Jacobs T, Assanis DN, Filipi Z. The impact of exhaust gas
recirculation onperformance and emissions of a heavy-duty diesel
engine. SAE technical paper2003-01-1068; 2003.
[45] Asad U, Kumar R, Zheng M, Tjong J. Ethanol-fueled low
temperaturecombustion: a pathway to clean and efficient diesel
engine cycles. ApplEnergy; n.d.
http://dx.doi.org/10.1016/j.apenergy.2015.01.057.
[46] Nakayama S, Fukuma T, Matsunaga A, Miyake T, Wakimoto T. A
new dynamiccombustion control method based on charge oxygen
concentration for dieselengines. SAE technical paper 2003-01-3181.
Warrendale, PA; 2003.
http://dx.doi.org/10.1016/j.pecs.2006.08.003http://dx.doi.org/10.1016/j.pecs.2006.08.003http://dx.doi.org/10.1016/j.fuel.2013.01.049http://dx.doi.org/10.1016/j.fuel.2013.01.049http://dx.doi.org/10.1016/j.fuel.2013.03.049http://dx.doi.org/10.1016/j.fuel.2013.03.049http://dx.doi.org/10.1016/j.enconman.2013.08.030http://dx.doi.org/10.1016/j.enconman.2013.08.030http://dx.doi.org/10.1016/j.biombioe.2013.03.004http://dx.doi.org/10.1016/j.biombioe.2013.03.004http://www.afdc.energy.gov/fuels/fuel_comparison_chart.pdfhttp://www.afdc.energy.gov/fuels/fuel_comparison_chart.pdfhttp://dx.doi.org/10.1023/B:IJOT.0000028470.49774.14http://dx.doi.org/10.1016/j.fuel.2009.08.043http://dx.doi.org/10.1016/j.measurement.2011.03.028http://dx.doi.org/10.1016/j.measurement.2011.03.028http://dx.doi.org/10.1243/09544070JAUTO935http://refhub.elsevier.com/S0196-8904(16)00059-5/h0170http://refhub.elsevier.com/S0196-8904(16)00059-5/h0170http://refhub.elsevier.com/S0196-8904(16)00059-5/h0170http://refhub.elsevier.com/S0196-8904(16)00059-5/h0175http://refhub.elsevier.com/S0196-8904(16)00059-5/h0175http://refhub.elsevier.com/S0196-8904(16)00059-5/h0175http://refhub.elsevier.com/S0196-8904(16)00059-5/h0180http://refhub.elsevier.com/S0196-8904(16)00059-5/h0180http://dx.doi.org/10.1016/j.apenergy.2014.02.073http://dx.doi.org/10.1016/j.apenergy.2014.02.073http://dx.doi.org/10.1016/j.enconman.2015.08.001http://refhub.elsevier.com/S0196-8904(16)00059-5/h0200http://refhub.elsevier.com/S0196-8904(16)00059-5/h0200http://refhub.elsevier.com/S0196-8904(16)00059-5/h0200http://dx.doi.org/10.1016/j.apenergy.2015.01.057
Energy efficiency impact of EGR on organizing clean
combustion�in diesel engines1 Introduction2 Experimental method3
EGR analysis3.1 Analytical approach3.2 Quantification of EGR
amount3.3 Intake dilution level3.4 Air–fuel ratio
considerations
4 EGR versus NOx and smoke emissions4.1 Single injection diesel
combustion4.2 Diesel injection with port fuelling
5 EGR and thermal efficiency5.1 Combustion phasing impact on
efficiency5.2 Combustion efficiency impact
6 EGR control considerations7
ConclusionsAcknowledgementsAppendix A EGR analysis for dual-fuel
modeReferences