混合动力汽车多目标改进型粒子群算法优化研究
Research on multi-objective optimization of hybrid vehicles based on improved particle swarm algorithm
邓涛 1马宝鹏 2谭孟骑3
作者信息
- 1. 重庆交通大学航空学院,重庆 400074;绿色航空能源动力重庆市重点实验室,重庆 400074;重庆交通大学绿色航空技术研究院,重庆 400074
- 2. 重庆交通大学交通运输学院,重庆 400074
- 3. 重庆交通大学机电与车辆工程学院,重庆 400074
- 折叠
摘要
为提高混合动力汽车的经济性、动力性和平顺性,以一款并联式混合动力汽车为对象,以控制策略参数和动力系统参数为优化变量,以动力电池荷电平衡等为约束条件,构建多目标优化模型.在优化过程中,引入混沌算子和余弦策略对粒子群优化算法的速度公式、惯性权重和学习因子进行改进,提出改进型粒子群优化算法,并进行仿真优化.结果表明,在满足约束条件的前提下,优化后经济性、平顺性和动力性分别提高了 15.88％、11.71％、3.51％.同时,发动机与电机工作点的效率分布得到明显改进.
Abstract
To improve the economy,power and smoothness of hybrid electric vehicles,we take a parallel hybrid electric vehicle as the research object and employ the control strategy parameters and power system parameters as optimization variables,and the power battery charge balance as constraints to build a Multi-objective optimization model.During the optimization process,chaos operators and cosine strategies are introduced to improve the speed formula,inertia weight and learning factor of the particle swarm optimization algorithm.We propose an improved particle swarm optimization algorithm with simulation and optimization.Our results show while meeting the constraints,our algorithm improves the economy,ride comfort and power performance by 15.88％,11.71％and 3.51％respectively after optimization.Meanwhile,the efficiency distribution of the engine and motor operating points improve markedly.
关键词
粒子群算法/多目标优化/混合动力汽车/Pareto最优解Key words
particle swarm optimization algorithm/multi-objective optimization/hybrid vehicles/Pareto optimal solution引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据