COMPARATIVE ANALYSIS OF MATHEMATICAL MODELS DEVELOPED BY RSM AND MLR METHOD FOR SFC OF SINGLE CYLINDER DIESEL ENGINE FUELED WITH DIESEL AND DIESEL-JATROPHA BLENDS

Abstract

The aim of the research is to compare the accuracy of multiple linear regression (MLR) model and response
surface method (RSM) model for specific fuel consumption (SFC) of a single cylinder diesel engine using various blends
of Jatropha biodiesel. Various researchers did an experiment to compare MLR and RSM method to check better method
for a better mathematical model, but no one did experiment related to the SFC comparison of a variable compression
ratio (VCR) single cylinder diesel engine. Compression ratio, biodiesel blend, injection pressure and load are the input
variables. The single cylinder diesel engine on which experiment performed is equipped with facility to vary compression
ratio, injection pressure and load. Three levels are selected for the parameters. Mathematical model was produced using
Response Surface Method in ‘Minitab 2017’ software for predicted SFC calculation. Then Multiple Linear Regression
carry out for SFC calculation. Comparison between MLR and RSM results were done. Diesel and jatropha biodiesel
blends was used as fuel in single cylinder diesel engine. It is found that the model developed using Response Surface
Methodology is more efficient and accurate than Multiple Linear Regression model. So RSM is more appropriate than
MLR.