• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2020, Vol. 56 ›› Issue (2): 98-105.doi: 10.3901/JME.2020.02.098

• 运载工程 • 上一篇    下一篇

基于多岛遗传算法的功率分流式双模混合动力客车参数优化

曾小华, 王振伟, 宋大凤, 陈琴琴, 杨南南   

  1. 吉林大学汽车仿真与控制国家重点实验室 长春 130025
  • 收稿日期:2019-01-09 修回日期:2019-09-15 出版日期:2020-01-20 发布日期:2020-03-11
  • 通讯作者: 宋大凤(通信作者),女,1977年出生,教授。主要研究方向为车辆地面力学与底盘电子集成控制。E-mail:songdf@126.com
  • 作者简介:曾小华,男,1977年出生,教授,博士研究生导师。主要研究方向为混合动力汽车。E-mail:zeng.xiaohua@126.com
  • 基金资助:
    国家重点研发计划资助项目(2018YFB0105900)。

Parameter Optimization of Dual-mode Power-split Hybrid Electric Bus Based on MIGA Algorithm

ZENG Xiaohua, WANG Zhenwei, SONG Dafeng, CHEN Qinqin, YANG Nannan   

  1. State Key Laboratory of Automotive Simulation and Control, Jilin University, Changchun 130025
  • Received:2019-01-09 Revised:2019-09-15 Online:2020-01-20 Published:2020-03-11

摘要: 协调优化混合动力系统设计参数和控制策略参数是提高功率分流式双模混合动力系统综合性能的关键。针对现有多目标遗传算法、粒子群算法以及模拟退火等优化算法实施难度高、求解周期长的问题,以汽车动力性能和电池SOC平衡为约束条件,将油电转换系数引入目标函数,提出一种基于多岛遗传算法(Multi-island genetic algorithm,MIGA)的动力系统设计参数和控制参数集成优化方法。利用Isight仿真平台进行建模仿真,结果表明,在保持车辆动力性、车速跟随以及电池SOC平衡的基础上,相较于优化前,在中国典型城市公交工况下系统等效油耗降低了0.21 L/100 km,对应节油比例为1.3%。对参数优化后的经济性仿真结果分析发现,优化后发动机向着高负荷区域移动,工作情况得到改善;电机MG1和MG2效率均有不同程度的提高,较好的验证了所提出的参数优化方法的有效性。

关键词: 车辆工程, 功率分流式双模混合动力系统, Isight仿真平台, 多岛遗传算法, 集成优化

Abstract: Coordinated optimization for parameters of hybrid electrical vehicle(HEV) powertrain and control strategy is the key to improving overall performance of power-split dual-mode hybrid electric bus. Aiming at the problems of high difficulty and long solution period for the optimization algorithms such as multi-objective genetic algorithm, particle swarm optimization and simulated annealing, the vehicle dynamic performance and battery SOC balance are taken as constraints, and the fuel-electricity conversion factor is introduced into the objective function. An integrated optimization method for design parameters and control parameters using multi-island genetic algorithm(MIGA) is proposed. The simulation have been carried out based on the Isight software, and the results show that on the premise of ensuring the power performance of vehicle and balance of battery SOC,the equivalent fuel consumption of the system is reduced by 0.21 L/100 km in the cycle of typical Chinese city, and the corresponding fuel saving ratio is 1.3%. The engine operating points move to the high load areas, and the working condition is improved. The efficiency of motor MG1 and MG2 is improved to different degrees, which demonstrated the effectiveness of the optimization algorithm in this work.

Key words: vehicle engineering, dual-mode power-split hybrid electric system, Isight software platform, multi-island genetic algorithm, integrated optimization

中图分类号: