Integrated optimization of common rail direct injection diesel engine input parameters with linseed biodiesel: A hybrid approach using grey relational analysis and genetic algorithm techniques
Abhishek Sharma,Nagendra Kumar Maurya,Avdhesh Tyagi,Nishant Kumar Singh,Yashvir Singh,Kaushalendra Kumar Singh,Manish Kumar
DOI: https://doi.org/10.1177/09544062241238797
2024-03-24
Proceedings of the Institution of Mechanical Engineers Part C Journal of Mechanical Engineering Science
Abstract:Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, Ahead of Print. In the pursuit of improved productivity and efficiency, the search for alternative fuels that reduce dependence on fossil fuels is of utmost importance. This research study aims to investigate the influence of process parameters, namely blend percentage, fuel injection pressure, exhaust gas recirculation (EGR) rate, and engine load, on various engine performance parameters such as torque, brake thermal efficiency (BTE), brake mean effective pressure (BMEP), hydrocarbon emissions (HC), and mechanical efficiency. The experiments are conducted following the Taguchi orthogonal array (L25) design of experiments. Additionally, grey relational analysis, a multi-objective optimization technique, is applied to identify the optimal levels of the process variables that optimize all the response parameters. The optimal values obtained through grey relational analysis are determined as 0% blend, 600 bar fuel injection pressure, 12% EGR rate, and 12 kg engine load. Moreover, the Genetic Algorithm-based Multi-Objective Genetic Algorithm (MOGA) is employed for multi-objective optimization of the engine input variables, yielding optimal levels of 19.45% blend, 594.35 bar fuel injection pressure, 14.12% EGR rate, and 12 kg engine load. Furthermore, multivariable regression models are developed to predict the response variables within the experimental domain. These models are validated through a confirmation test. The findings of this study provide insights into the optimization of process variables for enhanced engine performance, with the developed models serving as valuable tools for future predictions and optimization.
engineering, mechanical