摘要
针对传统遗传算法优化串联式混合动力汽车时燃油经济性和排放性不佳、优化运行时间过长的问题,提出一种基于特征选择遗传算法的串联式混合动力汽车系统参数和控制策略参数的多目标优化算法,并建立以动力性能指标为约束的混合动力汽车参数优化的非线性模型,其中目标函数包含最佳的燃油消耗和排放等指标。首先对目标函数的可行解空间进行特征选择,筛选出与目标函数相关度较高的可行解,并将基于特征选择的优化方法与传统遗传算法相结合,对其控制参数进行优化。其次通过ADVISOR仿真程序,计算出目标函数的最佳燃油消耗量和排放值。仿真结果表明:与传统的遗传算法相比,燃油经济性提高9.96%,CO、HC和NOX的排放分别降低18.07%、26.71%以及15.61%,同时优化运行时间降低62.14%。
Aiming at the problem of poor fuel economy and emission performance and long running time of optimized hybrid electric vehicle based on traditional genetic algorithm,In this paper,a multi-objective optimization algorithm for parameter and control strategy parameters of a series hybrid vehicle system with feature-based selection genetic algorithm is proposed,and a nonlinear model for hybrid vehicle parameter optimization based on dynamic performance index is established. The objective function contains the greatest indicators of fuel consumption and emissions. Firstly,the feature selection of the feasible solution space of the objective function is carried out,and the feasible solution with higher correlation with the objective function is selected. The optimization method feature-based selection is combined with the traditional genetic algorithm to optimize its control parameter. Secondly, through the ADVISOR simulation program, the optimal fuel consumption and emission value of the objective function are calculated. The simulation results show that compared with the traditional genetic algorithm,the fuel economy is improved by 9.96%,the emissions of CO,HC and NOxare reduced by 18.07%,26.71% and 15.61%,respectively,and the optimized running time is reduced by 62.14%.
作者
朱天军
胡伟
王林
ZHU Tian-jun;HU Wei;WANG Lin(College of Mechanical and Equipment Engineering,Hebei University of Engineering,Hebei Handan056038,China)
出处
《机械设计与制造》
北大核心
2020年第3期85-89,共5页
Machinery Design & Manufacture
基金
国家自然科学基金资助项目(51205105)
河北省高等学校科学技术研究项目(ZD2017213)
河北省科技计划项目(17394501D)
引进留学人员资助项目(CL201705)
河北省高层次人才资助项目(A2016002025)。
关键词
串联式混合动力汽车
多目标优化
目标函数
特征选择
控制参数
Series Hybrid Vehicle
Multi-Objective Optimization
Objective Function
Feature Selection
Control Parameter
