基于EMD与优化SVM的永磁同步电机失磁故障诊断研究  

Fault diagnosis method for PMSM demagnetization based on EMD and optimized SVM

作　　者：尹进田[1,2] 何志龙 刘丽[1,2] 邵武 YIN Jintian;HE Zhilong;LIU Li;SHAO Wu(Hunan Engineering Technology Research Center of Specialequipment ElectricEnergy Conversion and Control,Shaoyang University,Shaoyang 422000,Hunan,China;HunanProvincial Key Laboratory of Grids Operation and Controlon Multi-Power Sources Area,Shaoyang University,Shaoyang 422000,Hunan,China)

机构地区：[1]邵阳学院湖南省特种装备电能变换与控制工程技术研究中心,湖南邵阳422000 [2]邵阳学院多电源地区电网运行与控制湖南省重点实验室,湖南邵阳422000

出　　处：《农业装备与车辆工程》2024年第11期123-128,共6页Agricultural Equipment & Vehicle Engineering

基　　金：湖南省自然科学基金项目(2023JJ50270,2023JJ50267,2023JJ50263);湖南省教育厅重点项目(23A0543);湖南省科技计划项目(2016TP1023);湖南省2022年度教育科学规划课题(ND228480);邵阳学院研究生创新项目(CX2023SY054)。

摘　　要：针对永磁同步电机(PMSM)永磁体失磁故障初期难以及时发现的问题,提出一种基于经验模态分解(EMD)的故障诊断方法。采集PMSM的电流信号,通过EMD提取PMSM电流信号特征值,组合成数据集样本,构建出贝叶斯优化后的支持向量机(SVM)训练模型。将数据集输入优化SVM模型进行故障诊断,实验结果表明,使用贝叶斯优化后的SVM模型比传统SVM模型能够更加准确地实现PMSM失磁故障诊断。To address the issue of early detection difficulties in permanent magnet synchronous motor(PMSM)demagnetization faults,a fault diagnosis method based on Empirical Mode Decomposition(EMD)was proposed.Current signals of the PMSM were collected,and fault feature data samples were obtained through EMD.A Bayesian-optimized Support Vector Machine(SVM)training model was then constructed.The dataset was input into the optimized SVM model for fault diagnosis,and the model′s performance was validated using test data.Experimental results showed that,compared to traditional fault diagnosis methods,this algorithm could achieve more efficient and rapid diagnosis of PMSM demagnetization faults.

关 键 词：永磁同步电机 永磁体失磁 经验模态分解 贝叶斯优化 SVM模型 

分 类 号：TM341[电气工程—电机]