基于改进SDP-CNN方法的永磁轮毂电机退磁故障诊断研究

doi:10.3901/JME.260540

摘要/Abstract

摘要： 针对永磁轮毂电机故障诊断过程中数据预处理依赖人工经验与特征提取信息丢失问题，以永磁体退磁故障为载体，提出一种基于改进的对称点模式(Symmetric dot pattern,SDP)与卷积神经网络(Convolutional neural network,CNN)相结合的退磁故障诊断方法。首先，依据永磁材料退磁规律，建立考虑退磁故障的电磁场预测模型，结合仿真和试验获取不同退磁程度下的反电动势信号；其次，通过SDP将反电动势信号转换为雪花图像，并利用改进的鲸鱼优化算法(Whale optimization algorithm,WOA)对SDP方法中的参数进行自适应优化，实现故障特征的可视化增强并构建分类数据库；然后，通过改进的CNN算法进行特征提取与分类，建立故障信息与图像特征之间的映射关系，用于退磁故障的诊断。结果表明，所提出方法的故障诊断准确率达到99.3%，相较于传统CNN方法的准确率提升了1.7%，在噪声干扰下的诊断准确率仍能得到保证，能够有效适用于不同运行工况下的退磁故障诊断。

关键词: 永磁轮毂电机, 退磁故障诊断, 对称点模式, 鲸鱼优化算法, 卷积神经网络

Abstract: Data preprocessing relying on manual experience and the loss of information during feature extraction are two main challenges in the fault diagnosis of permanent-magnet in-wheel motors. A novel method is proposed for the diagnosis of permanent magnet demagnetization faults, which combines an improved symmetric dot pattern(SDP) with a convolutional neural network(CNN). First, based on the demagnetization characteristics of permanent magnet materials, an electromagnetic field prediction model is established to simulate motor operation under demagnetization fault conditions, and the back electromotive force signals are obtained from experiments. Next, the SDP method is employed to convert the back electromotive force signals into symmetrical snowflake images. To enhance the visualization of fault features, the parameters of SDP are adaptively optimized using an improved whale optimization algorithm(WOA), a classification database is established. Then, the database where feature extraction and classification are performed by using an improved CNN algorithm and a mapping relationship between fault information and image features is established for the diagnosis of demagnetization faults. The findings demonstrate that the fault diagnosis accuracy of the proposed approach has achieved 99.3%, marking a 1.7% enhancement compared to the conventional CNN method. Moreover, the diagnostic precision is maintained even under the influence of noise, ensuring its efficacy in diagnosing demagnetization faults across diverse operational environments.

Key words: permanent magnet in-wheel motor, demagnetization fault diagnosis, symmetrized dot pattern, whale optimization algorithm, convolutional neural network

中图分类号:

TG156

王强, 尚泽明, 王志勇, 任泽, 殷国栋. 基于改进SDP-CNN方法的永磁轮毂电机退磁故障诊断研究[J]. 机械工程学报, 2026, 62(14): 364-375.

WANG Qiang, SHANG Zeming, WANG Zhiyong, REN Ze, YIN Guodong. Demagnetization Fault Diagnosis Research of Permanent Magnet In-wheel Motor Based on Improved SDP-CNN Method[J]. Journal of Mechanical Engineering, 2026, 62(14): 364-375.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.cjmenet.com.cn/CN/10.3901/JME.260540

http://www.cjmenet.com.cn/CN/Y2026/V62/I14/364

参考文献

[1] 张奇祥，王金湘，张伊晗，等. 智能电动汽车线控制动关键技术与研究进展[J]. 机械工程学报，2024，60(10)：339-365. ZHANG Qixiang，WANG Jinxiang，ZHANG Yihan，et al. Key technologies and research progress of brake-by-wire system for intelligent electric vehicles[J]. Journal of Mechanical Engineering，2024，60(10)：339-365.
[2] 宋俊材，董菲，赵吉文，等. 基于重心邻域算法的无铁心永磁同步直线电机优化设计研究[J]. 中国电机工程学报，2017，37(12)：3594-3601. SONG Juncai，DONG Fei，ZHAO Jiwen，et al. Design optimization research of air-core permanent magnet permanent magnet synchronous linear motor based on gravity neighborhood center algorithm[J]. Proceedings of the Chinese Society of Electrical Engineering. 2017，37(12)：3594-3601.
[3] 梁修天，王其东，陈无畏，等. 基于功能安全的分布式驱动电动汽车前轮失效补偿控制[J]. 中国公路学报，2021，34(9)：26-38. LIANG Xiutian，WANG Qidong，CHEN Wuwei，et al. Compensation control of distributed drive electric vehicles based on functional safety[J]. China Journal of Highway and Transport，2021，34(9)：26-38.
[4] ABID A，KHAN M T，IQBAL J. A review on fault detection and diagnosis techniques：Basics and beyond[J]. Artificial Intelligence Review，2021，54(5)：3639-3664.
[5] FU Y，GAO Z，LIU Y，et al. Actuator and sensor fault classification for wind turbine systems based on fast fourier transform and uncorrelated multi-linear principal component analysis techniques[J]. Processes，2020，8(9)：1066.
[6] DE JESUS ROMERO-TRONCOSO R. Multirate signal processing to improve FFT-based analysis for detecting faults in induction motors[J]. IEEE Transactions on Industrial Informatics，2016，13(3)：1291-1300.
[7] 周小龙，刘薇娜，姜振海，等. 改进的HHT方法及其在旋转机械故障诊断中的应用[J]. 振动与冲击，2020，39(7)：189-195. ZHOU Xiaolong，LIU Weina，JIANG Zenhai，et al. Improved HHT method and its applications in rotating machinery fault diagnosis[J]. Journal of Vibration and Shock，2020，39(7)：189-195.
[8] LI Y，SONG W，WU F，et al. Spectral kurtosis of choi-williams distribution and hidden markov model for gearbox fault diagnosis[J]. Symmetry，2020，12(2)：285.
[9] ZHU X，HOU D，ZHOU P，et al. Rotor fault diagnosis using a convolutional neural network with symmetrized dot pattern images[J]. Measurement，2019，138：526-535.
[10] TRAN M Q，LIU M K，TRAN Q V，et al. Effective fault diagnosis based on wavelet and convolutional attention neural network for induction motors[J]. IEEE Transactions on Instrumentation and Measurement，2021，71：1-13.
[11] ZHU M，HU W，KAR N C. Torque-ripple-based interior permanent magnet synchronous machine rotor demagnetization fault detection and current regulation[J]. IEEE Transactions on Industry Applications，2016，53(3)：2795-2804.
[12] CUI W，MENG G，GOU T，et al. Intelligent rolling bearing fault diagnosis method using symmetrized dot pattern images and CBAM-DRN[J]. Sensors，2022，22(24)：9954.
[13] XU X，LIU H，ZHU H，et al. Fan fault diagnosis based on symmetrized dot pattern analysis and image matching[J]. Journal of Sound and Vibration，2016，374：297-311.
[14] XIA P，HUANG Y，TAO Z，et al. A digital twin-enhanced semi-supervised framework for motor fault diagnosis based on phase-contrastive current dot pattern[J]. Reliability Engineering & System Safety，2023，235：109256.
[15] SUN Y，LI S，WANG X. Bearing fault diagnosis based on EMD and improved chebyshev distance in SDP image[J]. Measurement，2021，176：109100.
[16] XIE J，ZHANG X，GUO J，et al. Fault detection of permanent magnet synchronous motor based on image content retrieval[J]. IEEE Transactions on Instrumentation and Measurement，2022，71：1-10.
[17] RAHMAN A，HE L，WANG H. Activation function optimization scheme for image classification[J]. Knowledge-Based Systems，2024：112502.
[18] HAO W，YIZHOU W，YAQIN L，et al. The role of activation function in CNN[C]//20202nd International Conference on Information Technology and Computer Application (ITCA). December 18-20，2020， Guangzhou，China. IEEE，2020：429-432.
[19] 王红霞，周家奇，辜承昊，等. 用于图像分类的卷积神经网络中激活函数的设计[J]. 浙江大学学报，2019，53(7)：1363-1373. WANG Hongxia，ZHOU Jiaqi，GU Chenghao，et al. Design of activation function in CNN for image classification[J]. Journal of Zhejiang University，2019，53(7)：1363-1373.
[20] 杜圣杰，贾晓芬，黄友锐，等. 面向CNN模型图像分类任务的高效激活函数设计[J]. 红外与激光工程，2022(3)：483-491. DU Shengjie，JIA Xiaofen，HUANG Yourui，et al. High efficient activation function design for CNN model image classification task[J]. Infrared and Laser Engineering，2022(3)：483-491.
[21] MULINDWA D B，DU S. An n-sigmoid activation function to improve the squeeze-and-excitation for 2D and 3D deep networks[J]. Electronics，2023，12(4)：911.
[22] ABDEL-NABI H，AL-NAYMAT G，ALI M Z，et al. HcLSH：A novel non-linear monotonic activation function for deep learning methods[J]. IEEE Access，2023，11：47794-47815.
[23] LIAO C，ZHANG Z，GU X，et al. A hybrid demagnetization performance prediction model for halbach array PMSM under symmetry short circuit fault[J]. IEEE Transactions on Industrial Electronics，2024，71(10)：12846-12857.
[24] KRICHEN M，ELBOUCHIKHI E，BENHADJ N，et al. Motor current signature analysis-based permanent magnet synchronous motor demagnetization characterization and detection [J]. Machines，2020，8(3)：35.
[25] NADERI P，FATHI A. Fault diagnosis/separation of surface mounted permanent magnet synchronous machine by current and its homopolar orders analysis[J]. IEEE Transactions on Energy Conversion，2022，38(2)：1246-1256.
[26] HUANG F，ZHANG X，QIN G，et al. Demagnetization fault diagnosis of permanent magnet synchronous motors using magnetic leakage signals[J]. IEEE Transactions on Industrial Informatics，2022，19(4)：6105-6116.
[27] SONG J，ZHAO J，DONG F，et al. A new demagnetization fault recognition and classification method for DPMSLM[J]. IEEE Transactions on Industrial Informatics，2019，16(3)：1559-1570.
[28] PARK Y，FERNANDEZ D，LEE S B，et al. Online detection of rotor eccentricity and demagnetization faults in PMSMs based on hall-effect field sensor measurements[J]. IEEE Transactions on Industry Applications，2018，55(3)：2499-2509.
[29] 金贤建，王佳栋，徐利伟，等. 轮毂电机驱动电动汽车主动悬架μ综合鲁棒控制研究[J]. 机械工程学报，2024，60(16)：259-269. JIN Xianjian，WANG Jiadong，XU Liwei，et al. μ-synthesis robust control for active suspension of in-wheel-motor-driven electric vehicles[J]. Journal of Mechanical Engineering，2024，60(16)：259-269.