基于ALIF-MPE-SVM组合算法的电机轴承早期故障诊断  

Early Fault Diagnosis of Motor Bearing Based on ALIF-MPE-SVM Combined Algorithm

作　　者：高美真[1] 李丽[1] 高烨童 薛涛[1] GAO Mei-zhen;LI Li;GAO Ye-tong;XUE Tao(School of Information Engineering,Jiaozuo Normal College,He’nan Jiaozuo 454000,China;School of Computer and Information Engineering,Xi’an University of Technology,Shaanxi Xi’an 710061,China)

机构地区：[1]焦作师范高等专科学校信息工程学院,河南焦作454000 [2]西安理工大学计算机与信息工程学院,陕西西安710061

出　　处：《机械设计与制造》2024年第8期202-205,211,共5页Machinery Design & Manufacture

基　　金：河南省高等学校重点科研项目(18B170004)。

摘　　要：为了提高电机用轴承的安全运行稳定效率,通过ALIF算法自适应分解非平稳信号,再以MPE从IMFs中提取出非线性故障信号,将MPE降维处理后的故障特征量利用MPE-SVM思想智能故障的诊断功能,开发得到一种MPE-SVM故障诊断技术,再根据测试得到的电机轴承故障参数完成算法有效性验证。研究结果表明:大部分故障信息都出现于最初的三个IMF内,主成分比例超过80%,因此以前3个主成分作为特征量并将其代入MPE-SVM内实施训练。各组别都可以对故障损伤的准确识别,表明以MPE作为故障特征能够满足有效性要求。ALIF-MPE具备比EMD-MPE更优的分类性能,达到了较低的标准差,稳定的分类状态。该研究能够准确识别电机轴承不同故障程度,对提高同类机械传动设备的故障诊断水平具有很好的理论支撑意义。In order to improve the safe operation and stability efficiency of motor bearings,non-stationary signals were decomposed adaptively by ALIF algorithm,and nonlinear fault signals were extracted from IMFs by MPE.The fault feature values after dimensionality reduction of MPE were used to develop a mPE-SVM fault diagnosis technology.Then the validity of the algorithm is verified according to the motor bearing fault parameters obtained by the test.The results show that most of the fault information occurs in the first three IMF,and the proportion of principal components exceeds 80%.Therefore,the former three principal components are used as characteristic quantities and substituted into MPE-SVM for training.Each group can accurately identify the fault damage,indicating that MPE as fault feature can meet the requirements of effectiveness.Alif-mpe has better classification performance than EMD-MPE,with lower standard deviation and stable classification state.This research can accurately identify different fault degrees of motor bearings,which has a good theoretical support for improving the fault diagnosis level of similar mechanical transmission equipment.

关 键 词：轴承 故障诊断 支持向量机 信息融合 特征提取 

分 类 号：TH16[机械工程—机械制造及自动化] TP277[自动化与计算机技术—检测技术与自动化装置]