CFD Optimization of the Pre-Chamber Geometry for a Gasoline Spark Ignition Engine

In the present paper, an efficient optimization method based on Bayesian updating strategy is developed for design of a spark-ignition engine equipped with pre-chamber. 3D computational fluid dynamics (CFD) simulation coupled strategies including experiment, genetic algorithm, and machine learning methods used to optimize pre-chamber desired combustion phasing. The process starts from experiment matrix 11 parameters, which are analytically characterize geometry set up CFD. Taking CA50 as single objective, CFD results then train models. Different models evaluated their Root Mean Square Error. Five five different categories selected second round evaluation. trained model in algorithm optimization, yields optimized configuration again justified by new added into database further refine model. After 24 iterations each models, medium Gaussian support vector identified best application. Iterations using continue until satisfactory result achieved. Detailed analysis conducted investigate physical mechanism about how influences engine's performance. It found that larger volume upper part stronger jet flow turbulent intensity accelerates flame propagation inside pre-chamber, dominating contrary effects reduced pressure temperature. Regression shows radius most influential parameter. current work not only sheds light design, but also has demonstrated general applicable purpose arbitrary mechanical system design.