汽车工程 ›› 2023, Vol. 45 ›› Issue (11): 2113-2122.doi: 10.19562/j.chinasae.qcgc.2023.11.012

所属专题: 新能源汽车技术-电驱动&能量管理2023年

• • 上一篇    下一篇

纯电动汽车传动系统-电机结构参数协同设计优化

何智成1,谢泽军1,刘侃1(),周恩临1,2,唐谦2,黄元毅3   

  1. 1.湖南大学,汽车车身先进设计制造国家重点实验室,长沙 410082
    2.湖南工程学院,汽车动力与传动系统湖南省重点实验室,湘潭 411014
    3.上汽通用五菱汽车股份有限公司技术中心,柳州 545000
  • 收稿日期:2023-03-29 修回日期:2023-05-21 出版日期:2023-11-25 发布日期:2023-11-27
  • 通讯作者: 刘侃 E-mail:lkan@hnu.edu.cn
  • 基金资助:
    国家重点研发计划(2022YFB3403202);国家自然科学基金联合基金(U20A20285);湖南省杰出青年基金(2021JJ10016)

Collaborative Design Optimization of Pure Electric Vehicle Drivetrain and Motor Structure Parameters

Zhicheng He1,Zejun Xie1,Kan Liu1(),Enlin Zhou1,2,Qian Tang2,Yuanyi Huang3   

  1. 1.Hunan University,State Key Laboratory of Advanced Design and Manufacture for Vehicle Body,Changsha  410082
    2.Hunan Institute of Engineering,Hunan Provincial Key Laboratory of Automotive Power and Transmission Systems,Xiangtan  411014
    3.SAIC-GM-Wuling Automobile Co. ,Ltd. ,Liuzhou  545000
  • Received:2023-03-29 Revised:2023-05-21 Online:2023-11-25 Published:2023-11-27
  • Contact: Kan Liu E-mail:lkan@hnu.edu.cn

摘要:

提出一种纯电动汽车传动系统与电机结构参数协同设计优化方法,来提高纯电动汽车动力性与经济性,同时降低永磁同步电机齿槽转矩以降低电机的振动噪声。首先,以电机结构参数为输入,额定转矩与齿槽转矩为输出,开展了基于电机多参数仿真和不同机器学习预测模型精度的研究,并建立永磁同步电机额定转矩和齿槽转矩的高精度机器学习预测模型;其次,利用电机基本设计参数(额定转矩、峰值转矩、额定转速、峰值转速)以及峰值效率构建永磁同步电机效率map图的快速预估数学模型;再次,基于电机齿槽转矩预测模型以及电机效率map图的快速预估数学模型,建立电机结构参数与效率特性的映射关系;最后,以电机结构参数和传动比为优化变量,整车动力性、经济性以及电机齿槽转矩为优化目标,运用遗传算法进行多目标优化。仿真结果表明,相较于优化前,优化后的整车性能0-100 km/h加速时间缩短了27.3%,15 km/h最大爬坡度提高了40.5%,WLTC工况能耗减少了1.6%,电机齿槽转矩降低了42.2%。

关键词: 永磁同步电机, 机器学习, 电机效率map图, 齿槽转矩

Abstract:

A collaborative design optimization method of pure electric vehicle drive train and motor structure parameters is proposed in this paper to improve the power and economy of pure electric vehicles, while reducing the cogging torque of permanent magnet synchronous motor to reduce vibration noise of motor. Firstly, with the motor structure parameters as input and rated torque and cogging torque as output, a study on the accuracy of multi-parameter simulation and different machine learning prediction models for motors is carried out, and a high-precision machine learning prediction model for rated torque and cogging torque of permanent magnet synchronous motors is established. Secondly, the basic motor design parameters (rated torque, peak torque, rated speed, peak speed) and peak efficiency are used to construct a mathematical model for fast prediction of the PM synchronous motor efficiency map. Thirdly, based on the motor cogging torque prediction model and the fast prediction mathematical model of motor efficiency map, the mapping relationship between the structural parameters of the motor and the efficiency characteristics is established. Finally, multi-objective optimization is carried out using the genetic algorithm with motor structure parameters and transmission ratio as optimization variables, and overall vehicle dynamics, economy and motor cogging torque as optimization objectives. Simulation results show that compared with that before optimization, the optimized vehicle performance is improved with 0-100 km/h acceleration time reduced by 27.3%, the maximum climbing degree of 15 km/h increased by 40.5%, the WLTC working condition energy consumption reduced by 1.6%, and the motor cogging torque reduced by 42.2%.

Key words: permanent magnet synchronous motor, machine Learning, motor efficiency map, cogging torque