Box–Behnken response surface methodology for optimization of operational parameters of compression ignition engine fuelled with a blend of diesel, biodiesel and diethyl ether

Abstract

The response surface method is adopted to express the output parameter (brake thermal efficiency) as a function of engine parameters including compression ratio (CR), injection pressure (IP) and injection timing (IT). The range of engine parameters selected is: 16 ≤ CR ≤ 18; 180 ≤ IP ≤ 220 bar; 15° ≤ IT ≤ 25⁰ BTDC. It is shown that the proposed Box–Behnken design required only 15 runs of experiments for the response surface between output parameters and engine parameters. Using ‘Design Expert’ software, 2D and 3D plots are generated. Such plots give an idea of domination of process variables and exhibit the trend of interaction between the variables. Moreover, a statistical significant test (ANOVA) is carried out to develop a ‘Regression Model’. The results of ANOVA suggested that the proposed ‘Regression Model’ is best fit with Standard Deviation of 0.093, Adjusted R² of 0.9619 and Adequate Precision of 19.686. Further the validation of ‘Regression Model’ is carried out by comparing the predicated values of thermal efficiency with that of experimental observations. The comparison suggested that the model is robust enough to predict thermal efficiency of the CI engine over a wide range of operating conditions with error less than 1%. Finally an optimum combination of engine parameters i.e. CR = 18, IP = 220 bar and IT = 20⁰ before top dead centre is proposed.

Keywords:

• RSM optimization
• CI engine
• Box–Behnken
• ANOVA
• brake thermal efficiency

Acknowledgements

The authors sincerely acknowledge the financial support received from Gujarat Council on Science and technology (GUJCOST), Gandhinagar, Gujarat, India, for the Minor Research Project entitled 'An Experimental Investigation on Life Cycle Analysis and Combustion Characteristics of CI Engine Operating on Esterified Oil and its Blends with Bio Additives’ wide letter No. GUJCOST/MRP/2014-15/392 dated 30/06/2014.

Disclosure statement

No potential conflict of interest was reported by the authors.