Multi-objective Optimization of Combustion Process in a Compressed Natural Gas Direct Injection Engine using Coupled Code of CFD and Genetic Algorithm 2007-01-1902

This study is concerned with the optimization of the combustion process in a compressed natural gas direct injection (CNG-DI) using the coupled CFD and multi-objective genetic algorithm (MOGA) code in order to improve the engine performance and emissions. The CFD algorithm was employed to simulate combustion reactions and the corresponding fluid dynamics while the genetic algorithm was utilized to optimize selected critical design parameters through a global search space. In this work, the CFD simulation took into account events such as injection of the gaseous fuel of natural gas directly into the combustion chamber along the central axis of the engine cylinder, the combustion reaction itself as well as emission formed as a result of the combustion. The aim of the MOGA optimization work was to obtain highest indicated power generation and least exhaust emissions produced. The coupled CFD-MOGA algorithm was used to investigate simultaneously the effects of three engine input parameters on the performance and emissions for a certain engine operating condition for which the baseline engine experimental data is available from experiments on the single cylinder research engine (SCRE). The start of injection (SOI), the end of injection (EOI) and the spark ignition (SI) timing are considered as important parameters which affected the engine performance in the form of indicated power due to combustion and exhaust emissions in the appearance of CO and NO as among the most hazardous emissions. The CFD-MOGA analysis was performed for homogeneous charge engine operation. The simulation result lead to identification of optimum timing of SOI, EOI and SI parameters for limited selected engine speeds, i.e. at 2000 rpm, which produced the highest engine power with and less exhaust emissions.