基于过程模拟技术和先进统计学方法的多参数发动机性能优化（英文）

Optimization of Multiple Parameters Engine Performance Based on Process Simulations and Advanced Statistics Methods

现代发动机在进行优化时参数较多,而且每个参数是多级别的,因此参数优化的工作量非常大,即使只在计算机上仿真优化也是如此.为了提高概念设计阶段的参数优化效率和精度,本文基于过程仿真模拟技术和先进的统计学方法对一台4缸4冲程、风冷、自然吸气发动机性能进行多参数优化研究.利用一维热力学仿真模拟软件GT-Power来模拟发动机性能,产生统计学分析所需要的基本数据.在模拟数据的基础上,利用实验设计方法(Do E)和响应面法(RSM)建立统计学模型对发动机性能进行优化.结果显示,RSM模型能准确预测发动机性能,并且通过该方法优化发动机的主要设计及运行参数,使发动机的最大输出功率提高了10%.

Many engine parameters need to be optimized in modern engine designing,and each parameter is with multiple levels,making the workload tremendous,even in computer simulation. To improve the parameters optimization efficiency and precision in the concept design phase,the process simulations and advanced statistic methods are coupled in this paper for parameter optimization of a 4-sroke,4-cylinder air cooled NA gasoline engine. The onedimensional thermodynamic simulation software GT-Power is used to generate the basic data used in the statistical analysis. A statistical model based on design of experiment（DoE） and response surface method（RSM） is then built to optimize the engine performance parameters. The results show that the RSM model can be used to accurately predict engine performance and the potential improvement for the maximum power of engine could be as high as 10%.