摘要
针对防爆电动机轴承运行信号采集时容易受到温湿度、尘埃、振动等因素干扰,常规故障诊断效果不佳的问题,提出并设计了一种基于改进SVM的矿用防爆电动机轴承故障诊断方法。首先提取矿用防爆电动机轴承故障特征,分析与处理电动机运行数据,得到轴承振动、声音、温度等特征参数。引入SVM理论,对其SVM核函数进行改进,完成防爆电动机轴承故障诊断模型的构建。采用粒子群算法完成模型求解,实现矿用防爆电动机轴承故障诊断。对比试验结果表明,应用该方法的诊断准确率更高,初步试验分析其具有一定可行性。
A fault diagnosis method for mining explosion-proof motor bearings based on improved SVM was proposed and designed to address the problem of poor conventional fault diagnosis results caused by interference from factors such as temperature,humidity,dust,and vibration during signal acquisition of explosion-proof motor bearings.Firstly,the fault characteristics of the explosion-proof motor bearings used in mining were extracted,the motor operation data were analyzed and processed,and characteristic parameters were obtained,such as bearing vibration,sound,temperature,etc.SVM theory was introduced and its SVM kernel function was improved to construct a fault diagnosis model for explosion-proof motor bearings.Particle swarm optimization algorithm was used to solve the model and achieve fault diagnosis of mining explosion-proof motor bearings.The comparative experimental results showed that the diagnostic accuracy of this method was higher,and preliminary experimental analysis showed that it had certain feasibility.
作者
王亦青
WANG Yiqing(CCRI(Beijing)Testing Technology Co.,Ltd.,Beijing 100013,China)
出处
《煤矿机电》
2025年第1期78-82,共5页
Colliery Mechanical & Electrical Technology
关键词
改进SVM
防爆电动机
轴承故障
诊断方法
improved SVM
explosion-proof electric motor
bearing fault
diagnostic method
