Comparative evaluation of various methodologies to account for the effect of load variation during cylinder pressure measurement of large scale two-stroke diesel engines D.T. Hountalas a,⇑ , R.G. Papagiannakis b , G. Zovanos a , A. Antonopoulos a a Internal Combustion Engines Laboratory, School of Mechanical Engineering, National Technical University of Athens, 9 Heroon Polytechniou St., Zografou Campus, 157 80 Athens, Greece b Thermodynamic & Propulsion Systems Section, Department of Aeronautical Sciences, Hellenic Air Force Academy, Dekelia Air Force Base, 1010 Dekelia, Attiki, Greece h i g h l i g h t s Use of cylinder pressure data for engine tuning leads to false results if load varies. Using two pressure sensors detects and accounts for the effect of load variation. Use of two pressure sensors is equivalent to simultaneous cylinder pressure measurement. Inlet pressure can be used to estimate load variation and correct engine performance data. A new computational method has been developed to account for load variation effect. a r t i c l e i n f o Article history: Received 22 May 2013 Received in revised form 14 August 2013 Accepted 17 August 2013 Available online 12 September 2013 Keywords: Diagnosis Engine condition monito ring Load variation Electric power station Diesel engine a b s t r a c t A sig nifi can t numb er of fa ult-de te cti on and fa ult dia gnosi s me thods ar e based on the use of the me asu re d cylinder pressure to estimate critical engine parameters i.e. cylinder brake power, fuel consumption, compression condition, injection timing etc. But, the results derived from the application of these tech- niques depend strongly on the quality of data used. A common problem which has been identified by many researchers is load variation during cylinder pressure measurement. This for some applications (marine) can become significant and in some cases makes unusable utilization of cylinder pressure mea- sure men t. Accordin g to the conv enti ona l mea sure men t tech niqu e for field appl icati ons, cyli nde r pres sure is measured sequ entially instea d of simultaneou sly due to issues related mainly to cost, applicability and complexity. Because of this, the operating parameters that are estimated for each cylinder depend on inst antaneous eng ine load . Th erefore wh en an oper atin g prob lem or a mistuning is iden tifie d, the disti nc- tio n for the actua l cau se (i. e. if it is att rib uted to a malfunct ion , mist uning or to engin e load variation dur- ing measurement), is difficult because cylinders are not measured simultaneously. For this reason, in the pr ese nt wo rk,threemeth od olo gie s that ha ve be en dev elo pe d to acc ount for the ef fec t of loa d va riation on diagnosis results are presented and evaluated in an attempt to be offered an alternative against simulta- neo us cyli nde r pre ssure me asur eme nt. For this pur pose , a wel l vali date d diag nostic tech niqu e, dev elop ed by the present research group, is employed and modified accordingly. The aforementioned methodolo- gies have been applied on a large-scale two-stroke diesel engine used for power generation in a Greek island at two different operating conditions. From the application of each method, diagnosis and tuning results are derived which are then compared against the respective ones obtained from the conventiona l diagnosis technique which neglects the effect of load variation during measurement. The evaluation is ba sedon thecomp ari sonof vit al en gin e pe rfo rmance data i.e . brake pow er outpu t, cy lin de r fuel flowrate , pea k firing pre ssure , ignition angle and compre ssion quality. From the comparison of the diagnos is results it is revealed that the three methodologies provide adequate results while the one which is based on the use of two cylind er pressure senso rs provides a very competitive alternativ e against simulta neous cylinder pressure measurement, offering the advantage of simplicity and low cost. Most important it is de mo ns tra ted the po ten tia l for al l three metho ds to propo se the req uired engin e tuning to guar antee un i- form cylinder operation despite variation of load during measurement. 2013 Elsevier Ltd. All rights reserved. 0306-2619/$ - see front matter 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apenergy.2013.08.036 ⇑ Corresponding author. Tel.: +30 210 772 1259; fax: +30 210 772 3475. E-mail addresses: [email protected](D.T. Hountalas), r.papagiannakis@ gmail.com(R.G. Papagiannakis). Applied Energy 113 (2014) 1027–1042 Contents lists available at ScienceDirect Applied Energy journal homepage: www.elsevier.com/locate/apenergy
16
Embed
Comparative Evaluation of Various Methodologies to Account
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
8/12/2019 Comparative Evaluation of Various Methodologies to Account
Comparative evaluation of various methodologies to account
for the effect of load variation during cylinder pressure
measurement of large scale two-stroke diesel engines
D.T. Hountalas a,⇑, R.G. Papagiannakis b, G. Zovanos a, A. Antonopoulos a
a Internal Combustion Engines Laboratory, School of Mechanical Engineering, National Technical University of Athens, 9 Heroon Polytechniou St., Zografou Campus,
157 80 Athens, Greeceb Thermodynamic & Propulsion Systems Section, Department of Aeronautical Sciences, Hellenic Air Force Academy, Dekelia Air Force Base, 1010 Dekelia, Attiki, Greece
h i g h l i g h t s
Use of cylinder pressure data for engine tuning leads to false results if load varies.
Using two pressure sensors detects and accounts for the effect of load variation.
Use of two pressure sensors is equivalent to simultaneous cylinder pressure measurement.
Inlet pressure can be used to estimate load variation and correct engine performance data.
A new computational method has been developed to account for load variation effect.
a r t i c l e i n f o
Article history:
Received 22 May 2013
Received in revised form 14 August 2013
Accepted 17 August 2013Available online 12 September 2013
Keywords:
Diagnosis
Engine condition monitoring
Load variation
Electric power station
Diesel engine
a b s t r a c t
A significant number of fault-detection and fault diagnosis methods are based on the use of the measured
cylinder pressure to estimate critical engine parameters i.e. cylinder brake power, fuel consumption,
compression condition, injection timing etc. But, the results derived from the application of these tech-
niques depend strongly on the quality of data used. A common problem which has been identified bymany researchers is load variation during cylinder pressure measurement. This for some applications
(marine) can become significant and in some cases makes unusable utilization of cylinder pressure mea-
surement. According to the conventional measurement technique for field applications, cylinder pressure
is measured sequentially instead of simultaneously due to issues related mainly to cost, applicability and
complexity. Because of this, the operating parameters that are estimated for each cylinder depend on
instantaneous engine load. Therefore when an operating problem or a mistuning is identified, the distinc-
tion for the actual cause (i.e. if it is attributed to a malfunction, mistuning or to engine load variation dur-
ing measurement), is difficult because cylinders are not measured simultaneously. For this reason, in the
present work, threemethodologies that have been developed to account for the effect of load variation on
diagnosis results are presented and evaluated in an attempt to be offered an alternative against simulta-
neous cylinder pressure measurement. For this purpose, a well validated diagnostic technique, developed
by the present research group, is employed and modified accordingly. The aforementioned methodolo-
gies have been applied on a large-scale two-stroke diesel engine used for power generation in a Greek
island at two different operating conditions. From the application of each method, diagnosis and tuning
results are derived which are then compared against the respective ones obtained from the conventional
diagnosis technique which neglects the effect of load variation during measurement. The evaluation is
basedon thecomparisonof vital engine performance data i.e. brake power output, cylinder fuel flowrate,
peak firing pressure, ignition angle and compression quality. From the comparison of the diagnosis
results it is revealed that the three methodologies provide adequate results while the one which is based
on the use of two cylinder pressure sensors provides a very competitive alternative against simultaneous
cylinder pressure measurement, offering the advantage of simplicity and low cost. Most important it is
demonstrated the potential for all three methods to propose the required engine tuning to guarantee uni-
form cylinder operation despite variation of load during measurement.
2013 Elsevier Ltd. All rights reserved.
0306-2619/$ - see front matter 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.apenergy.2013.08.036
The diesel engine remains the most efficient liquid fuel burning
unit yet devised and therefore it holds a dominant position in
many applications, i.e. marine propulsion, land transport – both
road and rail, power generation etc. [1–3]. Thus the proper and effi-
cient operation of a diesel engine is a major objective, especially for
marine and stationary applications (i.e. power generation etc.)
[4–6]. For this reason, condition monitoring and fault diagnosis
techniques hold an important position in the field especially for
the large-scale two-stroke diesel units due to their high power
output and fuel consumption [4–6]. The diagnosis procedure (i.e.
detection of the actual cause of a fault, engine mistuning etc.) is
usually very complicated, since engine performance is affected
from a large number of parameters, which are usually very difficult
to measure or estimate. Therefore the actual cause of a malfunction
cannot easily be determined using conventional methodologies. Up
to now, various diagnostic methods and techniques have been
proposed from researchers and manufacturers [7–10]. Most are
based on the processing of measurement data which are obtained
during engine operation [11–13]. A number of these techniques
make use of the measured cylinder pressure trace to estimate
critical engine parameters such as brake power, fuel consumption,
ignition angle etc. [14–16]. But for the acquired data to be reliable,
the measurement conditions must meet certain requirements
[7,17–19]. One of these which, is the subject of the present work,
is engine load which should remain constant during cylinder
pressure measurement [20–22]. According to the conventional
cylinder pressure measurement methodology, one pressure sensor
is used and cylinders are measured consequently one after the
other. The reason is that the simultaneous measurement requires
a great number of sensors, connections, sampling lines etc. that it
is not practical for field applications and moreover that this results
to high cost. Thus, if load variation occurs during cylinder pressure
measurement and is not accounted for, inaccurate results may be
derived for cylinder condition [20–22] and specifically for cylinder
tuning if these are directly utilized to adjust per example the fuel
flow (rack position) of individual cylinders. This will most possibly
result to engine mistuning with negative impact on engine perfor-
mance. Therefore it is important to precisely record load variation
during measurement.
As already mentioned one methodology to avoid the effect of
load variation during measurement is the simultaneous pressure
measurement of all cylinders, but for practical field applications,
it has various drawbacks. For this reason in the present work are
examined and evaluated three methodologies to detect record
and account for load variation during cylinder pressure measure-
ment. The first is based on the conventional cylinder pressure mea-
surement technique (i.e. cylinders are measured consequently one
after the other) and the simultaneous estimation of the charge
pressure corresponding to compression initiation, which is obvi-
ously affected by fluctuation of load. The second methodology is
based on the direct recording of the scavenging air pressure using
a fast response sensor with adequate accuracy, because the
Nomenclature
A area (m2)CF correction factor (–)CQ cylinder compression quality (%)CR compression ratio (–)c r radiation constant (W/m2 K4)
D cylinder bore (m) f number of cycles per secondhc heat transfer coefficient (W/m2 K)i cylinder number (–)l length (m)L percentage of full engine load (%)m mass (kg)_m mass flow rate (kg/s)
P pressure (N/m2)P e cylinder brake power output (W)P ind cylinder indicated power output (W)Q heat (J)t time (s)T temperature (K)V volume (m3)
X cylinders’ parameter (–)Y cylinder performance parameter (–) z number of cylinders
Dimensionless groupPr Prandtl number (–)Re Reynolds number (–)
AbbreviationsCA crank angledeg degreesLHV lower heating value (kJ/kg)rpm rotations per minuteSOI start of injection (deg CA)TDC top dead center (abbreviations)
1028 D.T. Hountalas et al. / Applied Energy 113 (2014) 1027–1042
In Fig. 9a–b is given the brake power output of the reference
cylinder, the actually measured cylinder power and the corrected
(i.e. estimated with 1st, 2nd or 3rd methodology) brake power out-
put of each cylinder, for 50% and 100% of full engine load at
143 rpm engine speed. Measured cylinder power is the one derived
from the application of the conventional diagnosis technique
which does not account for load variation (i.e. conventional brake
power). On the other hand, cylinder corrected power is the one de-
rived from the application of the three methodologies that are used
to account for the effect of load variation. Observing reference cyl-
inder brake power during measurement, it is revealed that for both
loads examined, a considerable fluctuation exists. This is an indica-
tion for engine load variation during the measurement. For this
reason, the measured cylinder power (i.e. the one derived through
the conventional methodology) has to be corrected properly, as al-ready described, in order the corrected power of each cylinder to
represent the actual power value, i.e. the one which would be re-
corded if engine load was kept constant during measurement.
Observing the corrected cylinder brake power output, esti-
mated from the first methodology, based on the variation of the
calculated cylinder charge pressure, it is shown that both the cor-
rected and the conventional brake power values follow a similar
variation trend among the cylinders. This becomes more evident
at full load condition. On the other hand, observing the corrected
cylinder brake power output estimated from the other two
0
20
40
60
80
100
120
140
C y l i n d e
r P r e s s u r e ( b a r )
50% Load
Mean Cycle
No1No2 No3 No4 No5 No6 No7
(a)0
20
40
60
80
100
120
140
160
180
C y l i n d e r P r e s s u r e ( b a r )
100% Load
Mean Cycle
No1 No2 No3 No4 No5 No6
No7
(b)
Fig. 7. Mean cylinder pressure diagrams at (a) 50% load and 143 rpm engine speed and (b) 100% load and 143 rpm engine speed.
Crank Angle (deg)
80
82
84
86
88
90
92
9496
98
C y l i n d e r P r e s s u r e ( b a r )
1st measurement
2nd measurement
3rd measurement
4th measurement
5th measurement
6th measurement
50% Load
Ref. cylinder
(a)
176 180 184 188 192 196
Crank Angle (deg)
120
122
124
126
128
130
132
134136
138
C y l i n d e r P r e s s u r e ( b a r )
1st measurement
2nd measurement
3rd measurement
4th measurement
5th measurement
6th measurement
100% Load
Ref. cylinder
(b)
176 180 184 188 192 196
Fig. 8. Cylinder pressure versus crank angle diagrams of the reference cylinder at (a) 50% load and 143 rpm engine speed and (b) 100% load and 143 rpm engine speed.
D.T. Hountalas et al./ Applied Energy 113 (2014) 1027–1042 1035
methodologies, based on the variation of the measured scavenging
pressure and the use of two cylinder pressure sensors, it is revealed
that both corrected brake power values follow an almost similar
variation trend among the cylinders but it is different compared
to the one of the first methodology. The difference is more obvious
at full engine load.
Using the cylinder brake power values reported in the previous
graphs are generated Fig. 10a–c, which provide the patterns of cyl-
inder power deviation around the mean value, for each methodol-
ogy at the two loading points examined. Power deviation of eachcylinder, on percentage basis, is estimated as follows:
DP e;ið%Þ; ¼ P e;i
P e;m
1
100 ð16Þ
where (P e,i) represents the corrected brake power of the (ith) cylin-
der and (P e,m) represents the mean value. Observing Fig. 10a, it re-
sults that, for the two load points examined, the power deviation
of each cylinder, derived from the first methodology is qualitatively
similar only for cylinders Nos. 1–3 and 5. On the other hand, this is
not the case for the remaining cylinders. Since the power deviationpattern for all cylinders should remain qualitatively the same with
Cylinder Number
850
900
950
1000
B r a k e P o w e r ( k W )
Measurement Number for Ref. Cylinder
50% LoadRef. cylinder
conv. method
1st method
2nd method
3rd method
(a)
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Cylinder Number
1550
1600
1650
1700
1750
1800
1850
B r a k e P o w e r ( k W )
0 1 2 3 4 5 6 0 1 2 3 4 5 6
Measurement Number for Ref. Cylinder
100% LoadRef. cylinder
Conv. method
1st method
2nd method
3rd method
(b)
Fig. 9. The actual and the estimated brake power output for each cylinder at (a) 50% load and 143 rpm engine speed and (b) 100% load and 143 rpm engine speed.
Cylinder Number
-6
-4
-2
0
2
4
6
( b r a k e p o w e r ) , ( % )
1st method
50 % Load
100 % Load
(a)Cylinder Number
-6
-4
-2
0
2
4
6 2nd method
50 % Load
100 % Load
(b)
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
0 1 2 3 4 5 6 7 8
Cylinder Number
-6
-4
-2
0
2
4
6 3rd method
50 % Load
100 % Load
(c)
( b r a k e p o w e r ) , ( % )
( b r a k e p o w e r ) , ( % )
Fig. 10. Comparisonof thedimensionless brake power output variationamong thecylinders for 50% and100%of full engine load, by using (a) the 1stdiagnosis methodology,
(b) the 2nd diagnosis methodology and (c) the 3rd diagnosis methodology.
1036 D.T. Hountalas et al. / Applied Energy 113 (2014) 1027–1042
8/12/2019 Comparative Evaluation of Various Methodologies to Account
engine load, this is an indication that the first methodology does not
totally detect and account for the effect of engine load variation.
Observing the results of the second methodology it is revealed
that cylinder power deviation pattern for cylinders Nos. 1–5 and
7 is similar for the two load points examined, while for No. 6 a dif-
ferent pattern is observed which could be attributed to a random
error during measurement. Moreover, comparing the absolute val-
ues of the power deviation of each cylinder for the two loading
points it is revealed that these are similar since only small differ-
ences exist.
Finally using the third methodology where load variation is di-
rectly detected form the power variation of the reference cylinder,
the patterns of power deviation all of cylinders appear to be similar
for both engine loading points examined, despite the small differ-
ences of absolute values attributed most possibly to measurement
error (normal).
Considering the previous, it is made obvious that the last two
methodologies adequately detect and account for the effect of en-
gine load variation on the cylinder pressure measurement and the
derived diagnosis results. Therefore, their advantage to properly
correct individual cylinder power considering for the effect of load
variation, is clearly demonstrated. However, it is to be noted that
the results of the first methodology can also be considered up to
a certain point as satisfying since the corrected results are close
to the reality. This is important because the specific methodology
is completely computational requiring not additional equipment
or measurement effort creating a strong motivation for further
investigation and improvement.
6.2.2. Cylinder tuning
Based on experience, cylinder power deviations in the range of
(±3%) are considered acceptable for the specific engine type. Differ-
ences above this range generate the necessity to adjust cylinder
power through fuel rack position adjustment. Considering the pre-
vious results two possibilities exist for cylinder tuning i.e. fuel rack
adjustment: Use of the primary measured cylinder brake power
(i.e. conventional brake power) and use of the corrected one which
estimated from the three methodologies examined in the present
work. As known in practice, the standard methodology is to use
the measured cylinder brake power output to analyze cylinder per-
formance. But, in the case of an engine load variation, as in the
present, there exists the risk for mistuning. For this reason in
Fig. 11a–b are given the estimates for cylinder brake power output
after having adjusted cylinder fuel racks using the corrected and
un-corrected vales of cylinder power output. It is evident that
the use of the uncorrected cylinder brake power output results to
significant distortion of engine operation in the case of a load
variation during the measurement. As shown, the first methodol-
ogy, based on the calculated of the cylinder charge pressure at
compression initiation would lead to a slight improvement, espe-
cially at low load but cannot guarantee uniform cylinder operation.
On the other hand the second methodology based on the measure-
ment of the instantaneous scavenging air pressure, results to
considerable improvement for both load points examined. How-
ever, the third methodology (i.e. the engine load variation is de-
tected through the brake power variation of the reference
cylinder) leads to uniform distribution of the brake power output
among the cylinders, clearly indicating that its advantageous and
capable to detect properly and eliminate accurately the effect of
engine load fluctuation during cylinder pressure measurement.
Consequently, considering the results provided in Fig. 11a and
b, it is obvious that the diagnosis technique should consider for
load variation during measurement, otherwise there is a risk for
engine mistuning. Finally, comparing the results it is revealed the
superiority of the third methodology for a correct cylinder power
tuning in the case of varying load during measurement.
6.2.3. Cylinder fuel flow rate
In Fig. 12a and b are given the corresponding values, corrected
and uncorrected, for each cylinder at 50% and 100% of full engine
load at 143 rpm engine speed. Cylinder fuel flow rate has been esti-
mated by applying the methodology described in Section 3.3 of the
present work. Since engine power is almost directly related to fuel
consumption, it is possible to correct cylinder fuel flow to account
for the effect of load variation using the three methodologies de-
scribed in the present work. Thus, in Fig. 12a and b is given the va-
lue for the corrected fuel flow rate of each cylinder which
corresponds to the value that would be observed if load had been
kept constant during measurement. The use of the conventional
cylinder fuel flow rate, without considering the load variation ef-
fect, for fuel rack adjustment will result to improper cylinder tun-
ing and most possibly to deterioration of engine performance.
Observing Fig. 12a–b it is revealed that for both loading points
examined, the comparison of the absolute values for the conven-
tional and corrected cylinder fuel consumption reveals that the
first methodology does not manage to account adequately for the
effect of load variation, since the results appear to be quite similar.
Moreover, the corrected fuel consumption, derived from the first
methodology, appears to follow a trend which is quite similar to
the respective one of the conventional methodology revealing its
difficulty to properly account for the effect of load variation.
On the other hand comparing the corrected fuel consumptions
of the remaining two methodologies, the results appear to be
rather similar. Both methodologies properly account for the effect
Cylinder Number
840
860
880
900
920
940
B r a k e p o w e r a f t e r t u n n i n g ( k W )
1st method
2nd method
3rd method
50% Load
(a)
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Cylinder Number
1620
1640
1660
1680
1700
1720
1740
B r a k e p o w e r a f t e r t u n n i n g ( k W )
1st method
2nd method
3rd method
100% Load
(b)
Fig. 11. Actual cylinder brake power versus estimated one after tuning using each one of the three methodologies, at (a) 50% load and 143 rpm engine speed and (b) 100%load and 143 rpm engine speed.
D.T. Hountalas et al./ Applied Energy 113 (2014) 1027–1042 1037
of load variation which is revealed from the fact that the corrected
fuel flow rates follow an almost similar variation trend among the
cylinders, which however is slightly different compared to the one
of the first methodology. The difference is more obvious at full en-
gine load.
Applying Eq. (16) for the corrected cylinder fuel consumption
are derived the graphs shown in Fig. 13a–c, which provide the cyl-
inder fuel consumption deviation around the mean value for both
load points examined.
Observing Fig. 13a, it results that, for both load points, the fuel
consumption deviation of each cylinder, derived from the first
methodology is qualitatively similar only for cylinders Nos. 1, 3,5 and 7. Furthermore, for both loads examined, the results ob-
tained from the second methodology provide the same pattern
for all cylinders except for the cylinder No. 6 where a slight differ-
ent pattern is observed. On the other hand, for both engine load
points, the pattern of the results obtained from the third method-
ology follow the same trend among all cylinders. Furthermore, very
small differences are observed between absolute values. The differ-
ence is almost the same for all cylinders revealing thus that it could
be attributed to experimental procedure.
Consequently, the second and third methodologies, based on
the use of the measured scavenging air pressure and the use of
two cylinder pressure sensors respectively, enable more accurate
estimation of the fuel flow rate to each cylinder compensatingfor the effect of load variation during the measurement. This is
Cylinder Number
160
165
170
175
180
F u e l C
o n s u m p t i o n ( k g / h )
Conv. method
1st method
2nd method
3rd method
50% Load
(a)
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Cylinder Number
275
285
295
305
315
F u e l C
o n s u m p t i o n ( k g / h )
Conv. method
1st method
2nd method
3rd method
100% Load
(b)
Fig. 12. The actual and the estimated fuel consumption for each cylinder at (a) 50% load and 143 rpm engine speed and (b) 100% load and 143 rpm engine speed.
Cylinder Number
-6
-4
-2
0
2
4
6 1st method
50 % Load
100 % Load
(a)
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
Cylinder Number
-6
-4
-2
0
2
4
6
( f u e l c o n s u m p t i o
n ) , ( % )
2nd method
50 % Load
100 % Load
(b)
0 1 2 3 4 5 6 7 8
Cylinder Number
-6
-4
-2
0
2
4
6 3rd method
50 % Load
100 % Load
(c)
( f u e l c o n s u m p t i o n ) ,
( % )
( f u e l c o n s u m p t i o n ) , ( % )
Fig. 13. Comparison of the dimensionless fuel consumption variation among the cylinders for 50% and 100% of full engine load, by using (a) the 1st diagnosis methodology,
(b) the 2nd diagnosis methodology and (c) the 3rd diagnosis methodology.
1038 D.T. Hountalas et al. / Applied Energy 113 (2014) 1027–1042
8/12/2019 Comparative Evaluation of Various Methodologies to Account
most important for tuning and cylinder power balancing because
fuel rack adjustment is made possible despite load variation during
the measurement.
6.2.4. Peak cylinder pressure
In Fig. 14a and b is given the peak cylinder pressure of the con-
ventional methodology and the corresponding corrected (i.e. esti-
mated with 1st, 2nd or 3rd methodology) values, which have
been derived from the application of all methodologies, for 50%
and 100% of full engine load at 143 rpm engine speed. From
Fig. 14a and b it results that, for both loads examined, the values
of corrected and un-corrected peak firing pressure follow the same
trend for all cylinders. Furthermore, comparing the absolute values
it results that the corrected ones derived from the first methodol-
ogy are almost the same to the ones of the conventional one. The
same conclusion is observed for the third methodology, exceptfor 50% of full load, where the corrected values for cylinders Nos.
3–5 and 7 appear to be slightly higher compared to the conven-
tional one. However, this could also be attributed to improper tun-
ing or to the function of the fuel injection system. On the other
hand it is revealed that, for both load conditions examined and
for all cylinders, the corrected values derived from the second
methodology are slightly lower compared to those of the other
two and the conventional methodology. However, it should be
Cylinder Number
90
94
98
102
106
P e a
k P r e s s u r e ( b a r )
Conv. method
1st method
2nd method
3rd method
50% Load
(a)
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Cylinder Number
130
134
138
142
146
150
P e a k P r e s s u r e ( b a r )
Conv. method
1st method
2nd method
3rd method
100% Load
(b)
Fig. 14. The actual and the estimated peak firing pressure for each cylinder at (a) 50% load and 143 rpm engine speed and (b) 100% load and 143 rpm engine speed.
Cylinder Number
-6
-4
-2
0
2
4
6
( p e a k p r e s s u r e ) ,
( % )
1st method50 % Load
100 % Load
(a)
Cylinder Number
-6
-4
-2
0
2
4
6 2nd method
50 % Load
100 % Load
(b)
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
0 1 2 3 4 5 6 7 8
Cylinder Number
-6
-4
-2
0
2
4
6 3rd method
50 % Load
100 % Load
(c)
( p e a k p r e s s u r e ) , ( % )
( p e a k p r e s s u r e ) , ( % )
Fig. 15. Comparison of the dimensionless peak firing pressure among the cylinders for 50% and 100% of full engine load, by using (a) the 1st diagnosis methodology, (b) the2nd diagnosis methodology and (c) the 3rd diagnosis methodology.
D.T. Hountalas et al./ Applied Energy 113 (2014) 1027–1042 1039
and mass leakage due to blow-by. This is achieved by the diagnos-
tic software [28,30]. Having determined all constant values, com-
pression quality is estimated from the comparison of the
effective compression ratio to the reference one (estimated from
the calibration procedure), as follows:
CQ ð%Þ ¼ CReff ;
curCReff ;ref
100 ð17Þ
where the effective compression ratio (CReff ) is the one definedfrom
the relation:
CReff ¼ P com
P o
1g
ð18Þ
where ‘‘P com’’ is the peak compression pressure estimated from the
diagnostic technique corresponding to the value if fuel flow was
instantaneously interrupted, ‘‘P o’’ is the cylinder charge pressure
at initiation of compression process, estimated from the diagnosis
procedure and ‘‘g’’ is the polytropic exponent. The reference effec-
tive compression ratio is obtained from the shop test data. In
Fig. 16a and b is given the compression quality derived from theapplication of the conventional cylinder pressure measurement
and the corrected values for each cylinder, for 50% and 100% of full
engine load at 143 rpm engine speed using the three methodolo-
gies. For constant load compression quality should remain almost
constant allowing only for the effect of load on blow-by since en-
gine speed is kept constant.
Observing the corrected values depicted in Fig. 16a and b it is
concluded that the first two methodologies, based on the estima-
tion or measurement of the instantaneous charge air pressure pro-
vide similar results for both load points examined. On the other
hand the third methodology provides results similar to the ones
Cylinder Number
80
84
88
92
96
100
C o m p r
e s s i o n Q u a l i t y ( % )
Conv. method
1st method
2nd method
3rd method
50% Load
(a)
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Cylinder Number
80
84
88
92
96
100
C o m p r e s s i o n Q u a l i t y ( % )
Conv. method
1st method
2nd method
3rd method
100% Load
(b)
Fig. 16. The actual and the estimated compression quality for each cylinder at (a) 50% load and 143 rpm engine speed and (b) 100% load and 143 rpm engine speed.
Cylinder Number
-0.8
-0.4
0.0
0.4
0.8
1.2
1.6
2.0
2.4
I g n i t i o n A n g l e ( d e g C A A T D C )
Conv. method
1st method
2nd method
3rd method
50% Load
(a)
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Cylinder Number
-0.8
-0.4
0.0
0.4
0.8
1.2
1.6
2.0
2.4
I g n i t i o n A n g l e ( d e g C A A T D C )
Conv. method
1st method
2nd method
3rd method
100% Load
(b)
Fig. 17. The actual and the estimated ignition angle for each cylinder at (a) 50% load and 143 rpm engine speed and (b) 100% load and 143 rpm engine speed.
1040 D.T. Hountalas et al. / Applied Energy 113 (2014) 1027–1042
8/12/2019 Comparative Evaluation of Various Methodologies to Account