Machine Learning Analysis Of CRDI Engine Performance Using Mahua Oil Blends Of Biodiesel

Abstract

This study examines the optimization of a Common Rail Direct Injection (CRDI) engine operating on a mahua oil and diesel blend to improve engine performance and minimize emissions. Response Surface Methodology (RSM) and Analysis of Variance (ANOVA) were utilized to evaluate the impacts of three principal input parameters: blend %, compression ratio, and injection pressure, including their interactions and higher-order effects. The findings indicated that compression ratio and injection pressure are critical parameters, exhibiting substantial nonlinear impacts on Brake Thermal Efficiency (BTE), Carbon Monoxide (CO), and Nitrogen Oxides (NOx) emissions. The blend %, while having a negligible direct effect, strongly interacted with other parameters to influence the engine's performance. The model demonstrated exceptional accuracy, with a R² value of 0.9952, signifying that it accounted for nearly 99% of the variability in the response. This study emphasizes the necessity of concurrently tweaking several engine parameters to attain peak efficiency and low emissions when utilizing biofuel blends. The results provide significant insights into the viability of mahua oil as a sustainable fuel for diesel engines, facilitating a transition to cleaner and more efficient biofuel use in contemporary engines.