考虑温度影响的车用驱动电机磁场分步泛化实时求解方法

doi:10.19595/j.cnki.1000-6753.tces.250947

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (7065 KB) HTML
输出: BibTeX | EndNote (RIS)

摘要准确快速量化温度对电磁性能的影响对电机的可靠运行、状态监测与优化设计具有重要意义。但是,车用永磁同步驱动电机工况复杂,工作温度动态多变,温度影响下电机励磁特性会发生变化,电机磁路也呈现高度非线性特征,传统有限元或热网络方法难以实现考虑温度的磁场快速准确求解。该文结合机器学习领域的迁移学习思想,提出一种在不同温度、不同电机工作点下分步泛化的磁场快速求解方法。首先采用本征正交分解对磁场矩阵进行降维,构建低维特征空间;然后结合迁移学习策略,将常温模型参数迁移至不同温度工况,显著减少训练数据需求;最后采用Kriging插值算法建立温度-工作点-磁场的映射关系,实现任意工况电机磁场的实时预测。结果表明,完成神经网络训练与插值后,所提方法实时计算完整电周期磁场的时间为0.95 s,相较于有限元求解方法,大幅降低了磁场求解所用时间;泛化范围内重构平均磁通密度相对误差不超过2.34%。通过搭载探测线圈的样机实验平台,验证了所提方法的有效性和准确性。提出的分步泛化策略为解决电机多场耦合中的温度-磁场的复杂交互关系提供了新的技术途径。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	王耀
	程远
	丁岭
	崔淑梅
	徐杰

关键词 ：永磁同步电机, 分步泛化, 迁移学习, 磁场计算

Abstract：Accurate and fast quantification of the effect of temperature on electromagnetic performance is of great significance for reliable operation, condition monitoring, and optimized design of motors. However, the working conditions of automotive permanent magnet synchronous drive motors are complex, and the working temperature is dynamic and variable. The motor's excitation characteristics change with temperature, and its magnetic circuit exhibits highly nonlinear behavior. It is difficult to obtain a fast, accurate solution for the magnetic field that accounts for temperature using traditional finite-element or thermal-network methods. This paper proposes a fast solution method for the magnetic field with stepwise generalization across different temperatures and motor operating points.
First, an electromagnetic simulation model of the motor accounting for temperature is established, and the magnetic field matrix is extracted from the simulation results. Secondly, a step-by-step generalization-solving method is proposed based on a migration-learning neural network and an interpolation algorithm. The magnetic field matrix is reduced using the POD algorithm to construct a low-dimensional feature space, and a neural network is trained to learn magnetic field features at different temperatures. The neural network and Kriging interpolation algorithm construct the mapping relationship of operating the point-temperature-magnetic field step by step. Finally, a prototype experimental platform is built to verify the proposed method.
The results show that the proposed method calculates the magnetic field over a complete electric cycle in 0.95 s, which is 1.12% of the time required by the finite element method. The relative error of the reconstructed average magnetic density in the generalized range does not exceed 2.34%. The proposed step-by-step generalization strategy provides a new way to solve the complex temperature-magnetic field interactions in multi-field coupling of electric machines.
The following conclusions can be drawn. (1) The proposed method has good generalization performance and meets the requirements of fast and accurate magnetic field reconstruction. (2) The proposed method effectively scales down the training cost of the neural network. Compared with a traditional data-driven neural network trained on 50% of the training set, the parameter freezing method further reduces network training time. The total training time of the network in the proposed method is 26.74% that of the neural network in the control group. (3) An experimental prototype with search coils is prepared, and the proposed method is verified.

Key words： Permanent magnet synchronous machine stepwise generalization transfer learning magnetic field computing

收稿日期: 2025-06-04

PACS:

TM351

基金资助:黑龙江省“双一流”新一轮建设学科协同创新成果建设资助项目（LJGXCG2022-065）

通讯作者: 丁岭男,1994年生,博士后,副研究员,研究方向为AI辅助的电机多物理场设计和优化、永磁同步电机稳健优化与高精度损耗分析、电机多目标智能优化、复杂机电系统多物理建模技术。E-mail: dingling@hitzri.cn

作者简介: 王耀男,1998年生,博士研究生,研究方向为电机数字孪生建模、高精度损耗分析、多物理场耦合及优化。

引用本文:

王耀, 程远, 丁岭, 崔淑梅, 徐杰. 考虑温度影响的车用驱动电机磁场分步泛化实时求解方法[J]. 电工技术学报, 2026, 41(10): 3273-3286. Wang Yao, Cheng Yuan, Ding Ling, Cui Shumei, Xu Jie. A Stepwise Generalized Real-Time Solution for the Magnetic Field of Automotive Drive Motors Considering Temperature Effects. Transactions of China Electrotechnical Society, 2026, 41(10): 3273-3286.

链接本文:

https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.250947 https://dgjsxb.ces-transaction.com/CN/Y2026/V41/I10/3273

[1] 王柄东, 王道涵, 王晓姬, 等. 交流调磁型永磁同步电机磁通协同调控最大转矩铜耗比控制[J]. 电工技术学报, 2024, 39(12): 3630-3645.
Wang Bingdong, Wang Daohan, Wang Xiaoji, et al.A maximum torque per copper loss control for AC flux-regulation permanent magnet synchronous motor with magnetic flux co-regulation[J]. Transactions of China Electrotechnical Society, 2024, 39(12): 3630-3645.
[2] 肖宗鑫, 胡明辉, 石力王, 等. 电动汽车内置式永磁同步电机转子温度在线估计[J]. 机械工程学报, 2023, 59(24): 209-222.
Xiao Zongxin, Hu Minghui, Shi Liwang, et al.Rotor temperature online estimation of built-in permanent magnet synchronous motor for electric vehicle[J]. Journal of Mechanical Engineering, 2023, 59(24): 209-222.
[3] 师蔚, 骆凯传, 张舟云. 基于热网络法的永磁电机温度在线估计[J]. 电工技术学报, 2023, 38(10): 2686-2697.
Shi Wei, Luo Kaichuan, Zhang Zhouyun.On-line temperature estimation of permanent magnet motor based on lumped parameter thermal network method[J]. Transactions of China Electrotechnical Society, 2023, 38(10): 2686-2697.
[4] 许孝卓, 郭国宾, 封海潮, 等. 基于代理模型的永磁直线同步电机多目标优化[J]. 电机与控制学报, 2024, 28(11): 139-150.
Xu Xiaozhuo, Guo Guobin, Feng Haichao, et al.Multi-objective optimization of permanent magnet linear synchronous motor based on surrogate model[J]. Electric Machines and Control, 2024, 28(11): 139-150.
[5] 李响, 郭鹏涛, 丁远. 基于场路结合的大功率直线超声波电机压电-热-结构多物理场分析[J]. 电工技术学报, 2024, 39(2): 423-433.
Li Xiang, Guo Pengtao, Ding Yuan.Piezo-thermal- structure coupling analysis for high-power linear ultrasonic motor based on field-circuit combination method[J]. Transactions of China Electrotechnical Society, 2024, 39(2): 423-433.
[6] 于占洋, 胡旭阳, 李岩, 等. 新型强迫风冷散热结构在高功率密度外转子表贴式PMSM上应用分析[J]. 电工技术学报, 2023, 38(24): 6668-6678.
Yu Zhanyang, Hu Xuyang, Li Yan, et al.Application analysis of novel forced air-cooled in outer rotor surface-mounted PMSM with high power density[J]. Transactions of China Electrotechnical Society, 2023, 38(24): 6668-6678.
[7] 陶飞, 张辰源, 戚庆林, 等. 数字孪生成熟度模型[J]. 计算机集成制造系统, 2022, 28(5): 1267-1281.
Tao Fei, Zhang Chenyuan, Qi Qinglin, et al.Digital twin maturity model[J]. Computer Integrated Manufa- cturing Systems, 2022, 28(5): 1267-1281.
[8] Tong Wenming, Sun Ruolan, Li Shiqi, et al.Loss and thermal analysis for high-speed amorphous metal PMSMs using 3-D electromagnetic-thermal bi- directional coupling[J]. IEEE Transactions on Energy Conversion, 2021, 36(4): 2839-2849.
[9] 卓亮, 孙鲁, 施道龙, 等. 考虑温度变化的高温高速永磁电机转子涡流损耗半解析模型及实验验证[J]. 中国电机工程学报, 2021, 41(24): 8305-8315.
Zhuo Liang, Sun Lu, Shi Daolong, et al.Semi- analytical model and experimental verification of rotor eddy current loss of high temperature high speed permanent magnet machine considering temperature change[J]. Proceedings of the CSEE, 2021, 41(24): 8305-8315.
[10] Cheng Yuan, Wang Yao, Sun Haodong, et al.Model order reduction based DQ axis parameter extraction method for IPMSMs[C]//2024 International Conference on Electrical Machines (ICEM), Torino, Italy, 2024: 1-7.
[11] Far M F, Martin F, Belahcen A, et al.Real-time control of an IPMSM using model order reduction[J]. IEEE Transactions on Industrial Electronics, 2021, 68(3): 2005-2014.
[12] 王耀, 程远, 高博, 等. 基于有限元降阶泛化的扁线绕组交流损耗快速计算方法[J]. 电工技术学报, 2025, 40(18): 5832-5844.
Wang Yao, Cheng Yuan, Gao Bo, et al.Fast calculation of AC losses in flat wire windings based on finite element reduced order generalization method[J]. Transactions of China Electrotechnical Society, 2025, 40(18): 5832-5844.
[13] 谢冰川, 张岳, 徐振耀, 等. 基于代理模型的电机多学科优化关键技术综述[J]. 电工技术学报, 2022, 37(20): 5117-5143.
Xie Bingchuan, Zhang Yue, Xu Zhenyao, et al.Review on multidisciplinary optimization key tech- nology of electrical machine based on surrogate models[J]. Transactions of China Electrotechnical Society, 2022, 37(20): 5117-5143.
[14] 朱新凯, 刘雅斌, 王景霞, 等. 基于非支配排序遗传算法与神经网络的20MW双定子超导磁场调制电机优化设计[J]. 中国电机工程学报, 2025, 45(15): 6103-6115.
Zhu Xinkai, Liu Yabin, Wang Jingxia, et al.Optimal design of 20MW double-stator superconducting magnetic field modulation electrical machine based on non-dominated sorting genetic algorithm and neural network[J]. Proceedings of the CSEE, 2025, 45(15): 6103-6115.
[15] 曹阳, 刘旭. 计及损耗的混合励磁电机建模与硬件在环实时仿真系统[J]. 电工技术学报, 2020, 35(22): 4657-4665.
Cao Yang, Liu Xu.Modeling method for hybrid- excited machine and hardware-in-loop real-time simulation system with accounting for loss calcu- lation[J]. Transactions of China Electrotechnical Society, 2020, 35(22): 4657-4665.
[16] 鲍明堃, 周扬忠. 基于复合神经网络重构对象的永磁同步直线电机变参数型位移速度并行控制[J]. 电工技术学报, 2024, 39(8): 2470-2484.
Bao Mingkun, Zhou Yangzhong.Parallel dis- placement velocity control of permanent magnet synchronous linear motor with variable parameters based on composite neural network reconstruction object[J]. Transactions of China Electrotechnical Society, 2024, 39(8): 2470-2484.
[17] Bellman R.Dynamic Programming[M]. Princeton Princeton: University Press, 1957.
[18] Ashraf M, Anowar F, Setu J H, et al.A survey on dimensionality reduction techniques for time-series data[J]. IEEE Access, 2023, 11: 42909-42923.
[19] Chu Shuaijun, Liang Deliang, Jia Shaofeng, et al.Research and analysis on design characteristics of high-speed permanent magnet claw pole motor with soft magnetic composite cores for wide temperature range[J]. IEEE Transactions on Industry Applications, 2022, 58(6): 7201-7213.
[20] Nielsen M A.Neural Networks and Deep Learning[M]. San Francisco: Determination Press, 2015.
[21] Yao Kai, Du Bochao, Li Jianhua, et al.Torque closed-loop flux-weakening control of IPMSM based on search coils[J]. IEEE Transactions on Industrial Electronics, 2025, 72(1): 122-133.