自動車工学の進歩
オープンアクセス

概要

Optimization of diesel engine parameters using genetic algorithms

V Magesh Kannan, Amirthagadeswaran, KS

This study is concerned with the selection of parameters of compression ignition engines, which affect fuel economy and harmful emissions such as carbon monoxide and nitrogen oxides. Experiments are conducted for performance and emissions by varying the operating parameters within the operating range. Full factorial experiments are conducted and this large amount of data is analyzed by a non-traditional soft computing technique, namely, Genetic Algorithm (GA). A mathematical model is formed using MINITAB software and the same model is used for optimizing the settings using GA. A single layer Levenberg-Marquardt backpropagation network is trained using the experimental data. Using the trained network, the output of the optimal parameter set obtained from the GA is predicted. The outputs from the experiments and GA are compared and the results are discussed. Applying this optimized parameter set to an engine will reduce harmful emissions of the engine, improve performance, save fuel and promote a cleaner environment.