基于SVM的无刷同步发电机故障检测研究  

Research on SVM-based Fault Detection for Brushless Synchronous Generators

作　　者：冯顺利[1] 饶美丽[2] 姜凌丽 FENG Shunli;RAO Meili;JIANG Lingli(School of Automobile,Henan College of Transportation,Zhengzhou 450005,China;School of Electronic and Electrical Engineering,Zhengzhou University of Science and Technology,Zhengzhou 450064,China)

机构地区：[1]河南交通职业技术学院汽车学院,河南郑州450005 [2]郑州科技学院电子与电气工程学院,河南郑州450064

出　　处：《机械制造与自动化》2023年第3期203-205,共3页Machine Building & Automation

基　　金：河南省教育厅自然科学基金项目(KJ201937875)。

摘　　要：旋转整流器是无刷同步电机中的重要部件,研究其故障检测方法对于提高发电机的整体运行水平十分必要。设计一种基于支持向量机(SVM)的无刷同步发电机旋转整流器故障检测方法,针对正常状态、整流器单个二极管开路状态以及整流器单个二极管短路状态,通过选取发电机三相端电压的多个特征频率,利用SVM进行特征分类,并借鉴Wrapper方法构造分类精度最高的频率特征子集,通过试验对所提方法进行验证。结果表明:由50 Hz(f_(2)特征)和150 Hz(f_(6)特征)构成的特征子集具有最高的分类精度,可用于旋转整流器的故障检测。As rotating rectifier is an important part of brushless synchronous generator,reserch on its fault detection method is of necessity for improvement of the overall operation level of the generator.A fault detection method based on support vector machine is proposed for the rotating rectifier of a synchronous generator.Aimed at the normal state,the single diode open state and the single diode short-circuit state of the rectifier,multiple characteristic frequencies of the three-phase terminal voltage of the generator are selected and SVM is applied for feature classifications.The frequency feature subset with the highest classification accuracy is constructed by Wrapper-based method,and the proposed method is verified by experiments.The results show that the feature subset composed of 50 Hz(feature f_(2))and 150 Hz(feature f_(6))has the highest classification accuracy and can be used for fault detection of rotating rectifier.

关 键 词：无刷同步电机 故障诊断 支持向量机 Wrapper方法 频率特征 

分 类 号：TM341[电气工程—电机]