|
|
|
| Fast Prediction of PWM Induced Losses for PMSM based on Subdomain Adaptation Transfer Learning Deep Neural Network |
| Wang Luyao1, Di Chong1, Guan Bokai1, Bao Xiaohua1, Li Shihao2 |
1. School of Electrical Engineering and Automation Hefei University of Technology Hefei 230009 China;
2. School of Electrical Engineering Hunan Mechanical & Electrical Polytechnic Changsha 410151 China |
|
|
|
Abstract Previous research has conventionally utilized the finite element analysis (FEA) to calculate losses of interior permanent magnet synchronous machines (IPMSM) under pulse width modulation (PWM) supply, which takes a lot of time to calculate the losses over the entire operating ranges. Semi-analytical methods and small-signal methods are proposed in some research to fasten the PWM voltage excitation (PVE) loss calculation. However, these approaches does not always have universality due to the complex structure of IPMSMs and software operation. To address this issue, this paper proposes a fast prediction method for PWM losses based on subdomain adaptive transfer learning deep neural network (SA-TLDNN).
Firstly, the losses under sinusoidal current exciton (SCE) and PVE are calculated to validate the similarity of the distribution under the two excitation modes. Besides, the SCE losses over entire operating range and PMW losses at several operating points are collected for training the model. The loss distribution is found to be complex and therefore, the subdomain adaptive method is utilized to divide the working points into several subdomains according to the control strategy. The loss distribution within each subdomain follows a regular pattern, which simplifies the model effectively. Aiming at the condition of small sample number of PMW losses, the gradient sequential sampling (GSS) method is utilized to fully capture the characteristics of PMW loss data. Based on the above methods, a SA-TLDNN model is established to maintain high prediction accuracy in the case of small samples. Finally, the model is trained using the previously prepared SCEFEA and PVEFEA data.
The root mean squared error (RMSE) is chosen to be the evaluation criteria. To validate the influence of sampling methods, the train progress is conducted under uniform sampling and GSS, respectively. The RMSE of the stator hysteresis loss is significantly decreased over 50% when GSS is applied. Subsequently, the operating range is divided into two subdomains to simplify the model. The RMSE of the stator hysteresis loss is reduced by 45.10% when the subdomain adaptation is conducted. The model is also trained with different sample numbers to validate the performance under few-shot condition. It is found that SA-TLDNN has the highest accuracy among other deep neural network models. The model-predicted results are compared with the FEA and experimental results. The relative error of PWM losses between predicted and simulated results are within 10%, which verifies that the proposed fast PWM loss calculation method has sufficient accuracy and can significantly reduce the calculation time.
The following conclusive are drawn from this research. (1) Transfer learning approach is well suited to the entire-range PWM loss calculation issue. (2) The subdomain adaption method can effectively reduce the model complexity and improve the prediction accuracy. (3) The source domain data feathers are effectively captured in few-shot condition. The loss prediction method based on SA-TLDNN adopted in this paper is able to realize the fast prediction of PWM losses under entire operating ranges, which can effectively facilitate the accurate and fast analysis of losses in the machine design process.
|
|
Received: 21 May 2025
|
|
|
|
|
|
[1] Xu Ran, Tong Wenming.Multi-objective hierarchical optimization of interior permanent magnet synchronous machines based on rotor surface modification[J]. CES Transactions on Electrical Machines and Systems, 2022, 6(4): 352-358.
[2] 肖宗鑫, 胡明辉, 周强, 等. 电驱系统轴承电蚀抑制措施研究综述[J]. 电工技术学报, 2025, 40(20): 6552-6576.
Xiao Zongxin, Hu Minghui, Zhou Qiang, et al.A review of research on bearing electrical-erosion suppression measures in electric drive systems[J]. Transactions of China Electrotechnical Society, 2025, 40(20): 6552-6576.
[3] Yamazaki K, Abe A.Loss investigation of interior permanent-magnet motors considering carrier harmonics and magnet eddy currents[J]. IEEE Transactions on Industry Applications, 2009, 45(2): 659-665.
[4] Yamazaki K, Seto Y.Iron loss analysis of interior permanent-magnet synchronous motors-variation of main loss factors due to driving condition[J]. IEEE Transactions on Industry Applications, 2006, 42(4): 1045-1052.
[5] Belahcen A, Arkkio A.Comprehensive dynamic loss model of electrical steel applied to FE simulation of electrical machines[J]. IEEE Transactions on Magnetics, 2008, 44(6): 886-889.
[6] Li Wei, Fan Xing, Zhu Lixun, et al.A dynamic hysteresis-based iron loss calculation method and its application in a PWM-fed IPMSM[J]. IEEE Transactions on Energy Conversion, 2024, 39(3): 1738-1749.
[7] Di Chong, Bao Xiaohua.A FEA-based fast AC steady-state algorithm for a voltage-driven PMSM by a novel voltage-flux-driven model[J]. IEEE Transactions on Industrial Electronics, 2024, 71(4): 3935-3943.
[8] Fan Xinggang, Li Dawei, Kong Wubin, et al.Fast calculation of strand eddy current loss in inverter-fed electrical machines[J]. IEEE Transactions on Industrial Electronics, 2023, 70(5): 4640-4650.
[9] 朱洒, 卢智鹏, 王卫东, 等. 基于CE-FEA和小信号分析快速计算逆变器供电下聚磁式场调制电机中永磁体涡流损耗[J]. 电工技术学报, 2020, 35(5): 963-971.
Zhu Sa, Lu Zhipeng, Wang Weidong, et al.Fast calculation of PM eddy current loss in FCFMPM machine under PWM VSI supply based on CE-FEA and small-signal analysis[J]. Transactions of China Electrotechnical Society, 2020, 35(5): 963-971.
[10] 朱洒, 曾峰, 陆剑波, 等. 考虑PWM谐波损耗的车用扁线内嵌式永磁同步电机效率图简化工程计算[J]. 电工技术学报, 2022, 37(22): 5687-5703.
Zhu Sa, Zeng Feng, Lu Jianbo, et al.Simplified engineering calculation of efficiency map of interior permanent magnet synchronous machines with hairpin windings considering PWM-induced harmonic losses[J]. Transactions of China Electrotechnical Society, 2022, 37(22): 5687-5703.
[11] Park S H, Kim K O, Lim M S.Computationally efficient estimation of PWM-induced iron loss of PMSM using deep transfer learning[J]. IEEE Transactions on Magnetics, 2023, 59(11): 8205305.
[12] Zhuang Fuzhen, Qi Zhiyuan, Duan Keyu, et al.A comprehensive survey on transfer learning[J]. Proceedings of the IEEE, 2021, 109(1): 43-76.
[13] Ma Yiming, Wang Jin, Xiao Yang, et al.Transfer learning-based surrogate-assisted design optimisation of a five-phase magnet-shaping PMSM[J]. IET Electric Power Applications, 2021, 15(10): 1281-1299.
[14] 谢冰川, 张岳, 刘光伟, 等. 基于小样本迁移代理模型的高速永磁电机应力优化[J]. 中国电机工程学报, 2024, 44(13): 5362-5374.
Xie Bingchuan, Zhang Yue, Liu Guangwei, et al.Stress optimization of high-speed permanent magnet motor based on few-shot transfer learning surrogate model[J]. Proceedings of the CSEE, 2024, 44(13): 5362-5374.
[15] 李泽泉, 马一鸣, 刘黎阳, 等. 基于几何相似性的迁移学习代理模型电机优化设计方法[J/OL]. 中国电机工程学报, 2025: 1-12. (2025-01-22). https://link.cnki.net/doi/10.13334/j.0258-8013.pcsee.242203.
Li Zequan, Ma Yiming, Liu Liyang, et al. Research on geometrical similarity transfer learning-based surrogate-assisted model optimization design method for electric machines[J/OL]. Proceedings of the CSEE, 2025: 1-12. (2025-01-22). https://link.cnki.net/doi/10.13334/j.0258-8013.pcsee.242203.
[16] Won Y J, Kim J H, Park S H, et al.Transfer learning-based design method for cogging torque reduction in PMSM with step-skew considering 3-D leakage flux[J]. IEEE Transactions on Magnetics, 2023, 59(11): 8204905.
[17] 王耀, 程远, 丁岭, 等. 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/OL]. 电工技术学报,1-14[2026-07-13].https://doi.org/10.19595/j.cnki.1000-6753.tces.250947.
Wang Yao, Cheng Yuan, Ding Ling, et al. A stepwise generalized real-time solution for magnetic field of automotive drive motors considering temperature effects[J]. Transactions of China Electrotechnical Society, [2026-07-13].https://doi.org/10.19595/j.cnki.1000-6753.tces.250947.
[18] 宋向金, 孙文举, 刘国海, 等. 深度子领域自适应网络电机滚动轴承跨工况故障诊断[J]. 电工技术学报, 2024, 39(1): 182-193.
Song Xiangjin, Sun Wenju, Liu Guohai, et al.Across working conditions fault diagnosis for motor rolling bearing based on deep subdomain adaption network[J]. Transactions of China Electrotechnical Society, 2024, 39(1): 182-193.
[19] Di Chong, Bao Xiaohua, Jiang Wei.Performance prediction of a ferrite-assisted synchronous reluctance machine considering different control strategies using open-source finite element analysis[J]. IEEE Transactions on Magnetics, 2023, 59(11): 8102905.
[20] Di Chong, Li Rongze, Wang Junhao, et al.A comprehensive study of a universal non-parametric mesh magnetic reluctance network model for radial interior permanent magnet synchronous machines[J]. IEEE Transactions on Energy Conversion, 2025, 40(1): 623-641.
[21] Zhao Wenxiang, Wang Yi, Ji Jinghua, et al.Reduced order model coupling frequency domain method for calculation of core loss in SPM machine considering PWM current harmonics[J]. IEEE Transactions on Industrial Electronics, 2025, 72(1): 220-229.
[22] Liao Chendong, Zhang Zhuoran, Han Jianbin, et al.Analysis and efficient calculation of the magnet eddy current loss of PMSM[J]. IEEE Transactions on Industry Applications, 2024, 60(6): 8797-8806.
[23] 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 technology of electrical machine based on surrogate models[J]. Transactions of China Electrotechnical Society, 2022, 37(20): 5117-5143.
[24] Xie Bingchuan, Zhang Yue, Xu Zhenyao, et al.Rotor multidisciplinary optimization of high speed PMSM based on multi-fidelity surrogate model and gradient sequential sampling[J]. IEEE Transactions on Energy Conversion, 2023, 38(2): 859-868.
[25] Pyrhonen J.Design of rotating electrical machines[M]. Weinheim: Wiley-VCH, 2013.
[26] Zhu Sa, Lu Jianbo, Zeng Feng.Fast calculation of electromagnetic forces in IPMSMs under PWM VSI supply based on small-signal time-harmonic finite element method[J]. CES Transactions on Electrical Machines and Systems, 2022, 6(1): 67-76. |
|
|
|
2026, Vol. 41 
