nd

Proceedings of the 2 International

INTERNATIONAL Conference on Current
JOURNAL Trends in Engineering and ENGINEERING
OF MECHANICAL Management ICCTEM -2014
17 19, July 2014, Mysore, Karnataka, India
AND TECHNOLOGY (IJMET)

ISSN 0976 6340 (Print)

ISSN 0976 6359 (Online) IJMET
Volume 5, Issue 9, September (2014), pp. 01-07
IAEME: www.iaeme.com/IJMET.asp
Journal Impact Factor (2014): 7.5377 (Calculated by GISI) IAEME
www.jifactor.com

OPTIMIZATION OF ENGINE OPERATING PARAMETERS

Nandkishore D.Rao1, Dr. B. Sudheer Prem Kumar2, Dr. C. Srinath3, Chandrashekar Patil4
1
Automobile Engg, Guru Nanak Dev Engineering College, Bidar, Mailoor Road- 585403, India
2
Mechanical Engg, JNT University College of Engineering, Hyderabad, India
3
Govt. Polytechnic for Women, Badang pet, Hyderabad, India
4
Mechanical Engg, Guru Nanak Dev Engineering College, Bidar, Mailoor Road, Bidar- 585403, Karnataka, India

ABSTRACT

This study is aimed at investigating the optimum combination of various engine operating parameters for
obtaining highest brake thermal efficiency and lowest emissions of smoke from a single cylinder diesel engine fuelled
with Honge oil- ethanol blend. During this work, the experiments are designed using a tool known as design of
experiments based on Taguchi and grey relational approaches. Engine operating parameters, namely, injector opening
pressure, fuel injection timing and compression ratio are varied at three levels and the responses like brake thermal
efficiency, brake specific energy consumption, emissions of oxides of nitrogen and smoke opacity are investigated. A
combination of injection pressure of 220 bar, injection timing of 27 before top dead centre and a compression ratio of 18
resulted in highest brake thermal efficiency and lowest smoke opacity. Results of confirmation tests indicated good
agreement with predicted quantities.

Keywords: Brake thermal efficiency, Design of experiment, Engine operating parameters, Grey relational approach,
Honge oil-ethanol blend.

1. INTRODUCTION

Energy is very essential for development of any country. With increasing industrialization and explosive growth
of vehicular population, the demand for energy is increasing at faster rate. To cater the energy demand, majority of
countries import petroleum products from oil rich countries this puts additional financial burden on their economy. In
year 2008, worlds total petroleum products resources were about 1238834 million barrels and the demand is estimated to
be 122 million barrels per day in 2032 [1-2].
With estimated petroleum reserves and projected demand, it is possible that the petroleum oil may deplete in
near future. Among the different petroleum products, consumption of diesel is highest because of widespread use of
compression ignition engines for transportation, power plants and agricultural implements, etc. due to their durability,
reliability and fuel economy. Combustion of diesel fuel leads to higher CO2 emission which is responsible for global
warming.
To overcome the problem of depletion of petroleum oil and to reduce the environmental pollution, it is
necessary to develop alternative fuels which are renewable, can be prepared locally and can emit low level of emittants to
air.
Among many alternatives, vegetable oils can be used as fuel for diesel engine due to their properties closer to
diesel and can be obtained from crops which ensure energy security. These are clean burning, renewable, non-toxic,

1
 Proceedings of the 2nd International Conference on Current Trends in Engineering and Management ICCTEM -2014
17 19, July 2014, Mysore, Karnataka, India

biodegradable, and environment friendly fuels that can be used in neat form or in blends with petroleum derived in diesel
engines.
Use of vegetable oils as fuel for diesel engine is known since its very first creation. But use of neat vegetable
oils indicates inferior engine performance due to their higher viscosity and lower volatility. To improve the fuel
properties, vegetable oils can be preheated, converted to biodiesel, blended with suitable solvents like diesel/alcohol, etc.
Preheating of vegetable oil requires additional heating arrangement which adds to the cost of engine. The trans-
estrification process (making biodiesel) requires costlier logistical support and skilled workforce. Mixing of vegetable oil
with diesel negates the idea of complete replacement of diesel. Blending of vegetable oil with oxygenated solvent like
ethanol can be done to reduce the viscosity for improvement of engine performance and to reduce the emissions like
hydrocarbon and smoke opacity due to presence of inherent oxygen in it. These blended vegetable oil fuels can be used
as complete replacement for diesel as these can be prepared at the site of application without any costlier logistical
support and skilled workforce.
As the fuel properties like viscosity, specific gravity and calorific value of vegetable oil- ethanol blends are not
same as that of diesel, it is essential to study the effect of variation of various engine operating parameters to optimize the
engine operation for highest brake thermal efficiency. The progress of combustion process depends on engine design
factors like shape and size of combustion chamber, location of nozzle and operating parameters like fuel injection
pressure, fuel injection timing, piston to head clearance, number and size of nozzle holes, etc. The design parameters are
set at design stage, it is not easy to change the design parameters as per type of fuel used, whereas it is easy to change the
operating parameters.
To find the optimum combination of above engine operating parameters for highest brake thermal efficiency
(BTE) and lowest smoke opacity (SO), variation in one-factor-at-a time method requires large number of trials, as there
could be many combinations of these parameters. Conduction of experimental investigations with all combinations of
engine operating parameters would be time consuming and costly.
Hence, some optimization approach can be used for finding suitable combination of engine operating parameters
so that engine can work with highest brake thermal efficiency and lowest emissions. The most common optimization
techniques which can be used for engine analysis are response surface method, grey relational analysis [3], non-linear
regression [4], genetic algorithm [5] and Taguchi method, etc.
In present investigation a blend of 70% Honge oil and 30% ethanol is used for operating diesel engine. The
Taguchi method along with gray relational analysis are used to study effect of several engine operating parameters
simultaneously with only few experiments to find optimum combination of these parameters for highest BTE and lowest
emissions.

2 TAGUCHI- GREY RELATIONAL APPROACH

The Taguchi with grey relational approach can be applied in following steps.

1. Identification of system output of interest and Selection of levels for the input factors.
2. Selection of appropriate orthogonal array (OA).
3. Conduction of experiments as per combinations of orthogonal array.
4. Normalization of system output values between zero and one and calculation of grey relational coefficient to
express relation between actual output values and desired values.
5. Calculation of overall grey relational grade using weightage factors.
6. Identification of level of each input factor with highest grey relation grade and S/N ratio (Signal to noise ratio).
7. Calculation of system output values at optimum combination of input parameters and Verification of responses
with confirmatory test under similar condition.

3 OPTIMIZATION OF ENGINE OPERATING PARAMETERS USING TAGUCHI AND GREY

RELATIONAL ANALYSIS

3.1 Selection of Engine Operating Parameters and Engine Responses

There are various engine operating parameters which affects the engine characteristics. Among those, in present
work, IOP, FIT and CR are considered for investigations. The different levels of IOP, FIT and CR considered in present
work are shown in Table no. 1. The engine responses selected are brake thermal efficiency (BTE), brake specific energy
consumption (BSEC), smoke opacity (SO) and oxides of nitrogen (NOx) emissions.

2
 Proceedings of the 2nd International Conference on Current Trends in Engineering and Management ICCTEM -2014
17 19, July 2014, Mysore, Karnataka, India

Details of engine operating parameters and their levels is shown in Table No. 1.

Table 1: Levels of engine operating parameters

Design factors Levels
1 2 3
Injector Opening pressure( bar) 200 220 240
Fuel injection timing (CA bTDC) 23 25 27
Compression ratio 17 17.5 18

3.2 Selection of Orthogonal Array

An orthogonal array presents the possible combinations of different levels of engine operating parameters which
are to be varied to investigate their effect on engine responses with smaller number of experimental trials. Selection of
orthogonal array depends on total degree of freedom of all engine operating parameters. For every engine parameter the
degree of freedom is calculated by using following relation
Degree of freedom for each engine operating parameter = (L-1) [6],
Where, L is the number of levels of each engine operating parameter
Total degree of freedom of all engine operating parameters,
N = (L-1)*P where. P= number of engine operating parameters

The OA must be selected such that, the number of trails in the selected orthogonal array (OA) must be equal to
the N+1[7].
As minimum number of trials to be conducted as per above relation is 7, an orthogonal array L9 (33) containing
9 trials has been selected in present investigation. The orthogonal array with different combinations of engine operating
parameters is shown in table 2.

Table 2: Details of combinations of engine operating parameters as per OA, L9.

Trial No. A (IOP) B (FIT) C ( CR)
1 200 23 17
2 200 25 17.5
3 200 27 18
4 220 23 17.5
5 220 25 18
6 220 27 17
7 240 23 18
8 240 25 17
9 240 27 17.5

3.3 Grey Relational Analysis of Experimental Data

The experiments are conducted as per the plan of L9 orthogonal array and the details of engine output responses for
BHO-70 are shown in table 3.

Table 3: Details of Engine responses for different trials

Trial No. BTE (%) BSEC (MJ/Kw-h) SO (%) NOx (ppm)
1 22.74 15.83 52 280
2 25.22 14.27 42 350
3 26.9 13.38 37 390
4 26.2 13.74 40 370
5 27.3 13.19 41 380
6 25.5 14.12 47 300
7 26.8 13.43 45 355
8 24.1 14.94 50 290
9 26.6 13.53 47 362

3
 Proceedings of the 2nd International Conference on Current Trends in Engineering and Management ICCTEM -2014
17 19, July 2014, Mysore, Karnataka, India

From these experimental results, the normalized data and grey relational coefficient for every engine response
are calculated.
The normalized data for minimized responses like BSEC, NOx and SO corresponding to lower the better criteria
is calculated using

max ai(k ) ai(k )

xi (k) =
max ai(k ) min ai(k )

Normalised data for maximised response like brake BTE corresponding to higher-the-better is calculated by

ai(k ) min ai(k )

xi(k)=
max ai(k ) min ai(k )

Where xi(k) is the normalised value of i-th response for k-th trial after the grey relational generation, ai(k ) is the
value of i-th response for k-th trial, min ai (k ) is the smallest value of ai (k ) , and the max ai (k ) is the largest value of
the ai (k ) [8].

The grey relational coefficient i (k) for every response at every trial is calculated by using relation

min + max
i (k) =
oi(k ) + max

where,

i(k)=grey relational coefficient, oi = xo ( k ) xi ( k ) =difference of between ideal normalised value xo(k) ( i.e. 1)
and xi(k) (normalised value of i-th response for k-th trial) [7]. min and max are the minimum and maximum values of the
oi of all trials. is a distinguishing coefficient, 0 1[8].

Grey relational co-efficient for BTE, BSEC, NOx and SO at every trial are calculated and shown in table no.4.
After calculating the grey relational coefficients, by selecting appropriate weighting factor i for every engine
response (can be specified from experience and are given in table no. 5.), the grey relational grade k for every trial is
calculated by using following relation and values of grey relational grade as shown in table 6.

n
k = i(k )i
k =1

Where, i = 1 and n- number of engine responses and i(k ) is grey relational coefficient
Table 4: Grey relational coefficient for engine output responses

Trial No. BTE BSEC SO NOx

1 0.33333 0.33333 0.333 1
2 0.52293 0.54867 0.6 0.44
3 0.85074 0.87084 1 0.33333
4 0.67455 0.70484 0.714 0.37931
5 1 1 0.652 0.35483
6 0.55882 0.58684 0.429 0.73333
7 0.82014 0.84311 0.484 0.42307
8 0.41605 0.43023 0.366 0.84615
9 0.76510 0.79210 0.429 0.40145

4
 Proceedings of the 2nd International Conference on Current Trends in Engineering and Management ICCTEM -2014
17 19, July 2014, Mysore, Karnataka, India

Table 5: Weighting factors for engine responses

Response factor Weighting factor
BTE 0.5
BSFC 0.1
SO 0.3
NOx 0.1

Table 6: Grey relational grade and S/N ratio

Trial. No. Grey relational Grade S/N Ratio

1 0.4 -7.958800173
2 0.54033541 -5.346731415
3 0.84579115 -1.45473727
4 0.659979811 -3.609386989
5 0.831136045 -1.606557654
6 0.54000112 -5.352106794
7 0.68185254 -3.326190743
8 0.445424263 -7.024522602
9 0.630477843 -4.006603429
Average 0.619444242 -7.958800173

Then average grey relational grade for every level of engine operating parameter is calculated by averaging the
values of grey relational grade for trials with given level of engine parameter.
Higher value of grey relational grade is considered as the stronger relational degree between the ideal level of
engine operating parameter and the given level of operating parameter. Thus, the higher relational grade implies that the
corresponding engine parameter combination is closer to the optimal. The grey relation grade with weighting factor is
analyzed by S/N ratio. The S/N ratio is selected as larger the better since the higher value of grey relational grade shows
the closeness of ideal and the given combination of engine parameters [8].

The Signal to Noise ratios (S/N)

For system responses Smaller-The-Better:

S/N = -10 Log10 [mean of sum of squares of measured data]

This type of S/N ratio is chosen for undesirable system outputs like emissions, fuel consumption, etc. for which
ideal value is zero.

For system responses Larger-The-Better:

S/N = -10 Log10 [mean of sum squares of reciprocal of measured data]

This case has been converted to smaller-the-better by taking the reciprocal of measured data and then taking the
S/N ratio as in the smaller-the-better case.
An average grey relational grade and average signal to noise ratio for each engine parameter is shown in Fig. 1
and Fig. 2 respectively.

5
 Proceedings of the 2nd International Conference on Current Trends in Engineering and Management ICCTEM -2014
17 19, July 2014, Mysore, Karnataka, India

Fig. 1: Mean S/N at different engine operating parameters

Fig. 2: Mean Grey relational Grade at different engine operating parameters

From above figures, it is observed that the compression ratio has significant effect on engine responses (as the
slope of mean grey relational grade or S/N ratio is observed to be highest for compression ratio)
The optimum combination of engine operating parameters is observed to be A2B3C3, i.e. at 220 bar, 27 bTDC
and 18 as at these levels highest S/N ratio or mean grey relational grade is observed
The S/N ratio or grey relation grade at optimum level of process parameters can be calculated as:

o
opt = mean + (ol mean)
i =1

Where opt is the mean S/N ratio or mean grey relation grade for optimum combination of engine parameters,
mean is the mean of the grey relation grade for all trials, ol is the value of mean grey relation grade at optimum
level of each engine operating parameter and o is the number of the engine parameters that affect the engine responses.
Table 7 shows Comparison of grey relational grade and S/N ratio of initial condn with optimum condn.

The values of engine responses at optimum combination of engine operating parameters are calculated by
o
OPT = T + ( Xi T ) [7]
i =1
Where, T is the overall mean value of the output response variable for the test runs conducted and Xi mean of the
engine response for the trials with optimum level of the engine operating parameter X.

6
 Proceedings of the 2nd International Conference on Current Trends in Engineering and Management ICCTEM -2014
17 19, July 2014, Mysore, Karnataka, India

3.4 Confirmation Test

To compare the engine characteristics at optimum combination of engine operating parameters, a confirmation
test is conducted under similar condition. Table no.8 shows the comparison of engine responses at initial engine settings,
predicted and experimental responses at optimized conditions. An improvement in engine output response at optimized
condition of 220 bar-27bTDC and 18 CR is observed as compared to initial conditions.

Table 7: Comparison of grey relational grade at initial condition with optimum condition
Initial Parameter
Combination Prediction Confirmation Test
Level A1B1C2 A2B3C3
Grey relational Grade 0.610264 0.896500 0.8956
S/N ratio -4.289639 -0.9489 -0.957

Table 8: Comparison of engine responses at initial engine settings, predicted and experimental responses at
optimized conditions
Sl. No Response Value of the response variables
variables
Initial Condition Predicted Conf. test
(T-G )
1 BTE 24.6 28.34 26.85
2 BSEC 14.63 12.51 13.40
3 SO 46 38.22 39
4 NOx 285 391.88 395

4. CONCLUSION

The Taguchi approach along with grey relational analysis has been used for optimizing the performance of
diesel engine fuelled with blend of honge oil and ethanol. The CR was found to be the most significant parameter. Based
on this study, it can be concluded that BTHE, BSEC, and emissions of diesel engine depend upon CR, injector opening
pressure and fuel injection timing. It is found that a diesel engine operating at a CR 18, pressure 220 bar, IT of 27
bTDC, achieves the optimum engine performance. The calculated results are well supported by the findings of
confirmatory test.

5. REFERENCES

[1] 2010 survey of energy resource, world Energy Council, 2010.

[2] Basic Statistics on Indian Petroleum and Natural Gas Ministry of Petroleum and Natural Gas, Government of
India, 2010.
[3] Karnwal, A., Multi- Response Optimization of Diesel Engine Performance Parameters Using Thumba Biodiesel
Diesel Blends by Applying the Taguchi Method and Grey Relational Analysis, International Journal of
Automotive Technology,12(4), 2011, 599-610.
[4] Maheshwari, N., A Nonlinear Regression Based Multi-Objective Optimization of Parameters Based on
Experimental Data from an IC Engine Fueled with Biodiesel Blends, Biomass and Bioenergy, 35(5), 2011,
2171-2183.
[5] Alonso, J. M., Combining Neural Networks and Genetic Algorithms to Predict and Reduce Diesel Engine
Emission, IEEE Trans E., 11(1), 2007, 46-55.
[6] Phillip J. Ross., Taguchi Techniques for Quality Engineering, (McGraw-Hill Book Company New York, 2002).
[7] S. Kaliamoorthy and Ravikumar Paramasivam, Investigation on performance and emissions of a biodiesel engine
through optimization techniques, Thermal science, 17(1), 2013, 179-193.
[8] Ambarish Datta, Optimization of Engine Performance and Emission Characteristics of Variable Compression
Ratio Diesel Engine Fuelled with Karanja Oil Methyl Ester using Taguchi Method and Grey Relational Analysis,
Jadavpur university M. Tech. Thesis, 2011.