Chapter 6 - A machine learning-genetic algorithm approach for rapid optimization of internal combustion engines

Abstract

The optimization of internal combustion (IC) engine is a highly complex problem because of the high-dimensionality and nonlinear interactions among design parameters. Machine learning (ML) offers a pathway to transform complex physical processes that occur in a combustion engine into compact informational processes. A careful definition of the objective (merit) function for optimization is a critical step. Training data for these ML algorithms must be cleverly prepared to improve the prediction efficiency. Global optimum search optimization methods must be adopted to avoid local optimum designs with reduced merit value. Postprocessing of the optimization outputs is also needed to evaluate the recommended design by employing sensitivity and robustness analysis.