基于多域特征的BA - KELM微型电机故障检测  

Fault detection of micro motors based on the BA - KELM utilizing multi-domain features

作　　者：郭明军 李伟光[2] 赵学智[2] 张欣欣 GUO Mingjun;LI Weiguang;ZHAO Xuezhi;ZHANG Xinxin(School of Mechanical and Automotive Engineering,Guangxi University of Science and Technology,Liuzhou 545616,China;School of Mechanical&Automotive Engineering,South China University of Technology,Guangzhou 510640,China)

机构地区：[1]广西科技大学机械与汽车工程学院,广西柳州545616 [2]华南理工大学机械与汽车工程学院,广州510640

出　　处：《振动与冲击》2023年第2期251-257,共7页Journal of Vibration and Shock

基　　金：国家自然科学基金(51875205);广西科技大学博士基金项目(校科博21z59)

摘　　要：针对目前对于微型电机故障检测的研究较少及基于单域特征的传统电机诊断方法精度低等问题,提出一种基于集成经验模态分解(ensemble empirical mode decomposition, EEMD)及蝙蝠算法(bat algorithm, BA)优化核极限学习机(kernel based extreme learning machine, KELM)的微型电机故障检测方法。所提方法包括样本集构造、模型训练及参数优化和模型测试3个步骤:首先,对采集到的微型电机信号进行EEMD处理并依据相关系数原则筛选出主要的本征模态分量(intrinsic mode fuction, IMF),结合计算得到的电机信号的时域、频域特征构造多域特征集并进行归一化处理,按一定比例将样本集划分训练集和测试集;其次,输入训练集,以错误率为适应度,并采用蝙蝠算法对KELM模型进行参数优化;最后,输入测试集对优化的BA-KELM模型进行测试,并与其他模型进行对比。试验结果表明,所提方法的准确率达98.75%,高于其余方法。At present, there are few researches on micro-motor fault detection, and the traditional motor diagnosis methods based on single domain features have usually low accuracy. So, a fault detection method for micro motors based on the ensemble empirical mode decomposition(EEMD) and bat algorithm(BA) was proposed. The proposed method includes three steps: constructing sample sets, model training as well as parameters optimizing, and model testing. Firstly, EEMD processing was carried out on the collected micro motor signals, and the main intrinsic mode fuction(IMF) components were selected by virtue of the principle of correlation coefficient. Fusing the calculated time and frequency domain features of the motor signals, a multi-domain feature set was constructed and normalized. Then, these features were divided into a training set and a test set according to a certain proportion. Taking the training set as input and employing, the error rate as fitness, the parameters of the kernel based extreme learning machine(KELM) model were optimized by means of the BA. Finally, the optimized BA-KELM model was tested by using the test set. The experimental results show that the accuracy of the proposed method is 98.75%, which is higher than other methods.

关 键 词：微型电机 故障诊断 蝙蝠算法(BA) 极限学习机(ELM) 核极限学习机(KELM) 

分 类 号：TH17[机械工程—机械制造及自动化]