FBF网络和小波包分析在机车轴承故障诊断中的应用  

Application of FBF Network and Wavelet Packet Analysis in Locomotive Bearing Fault Diagnosis

作　　者：王贵鑫 WANG Guixin(Testing and Rescue Branch of Guoneng Shuohuang Railway Development Co.,Ltd,Cangzhou,Hebei 062350)

机构地区：[1]国能朔黄铁路发展有限责任公司检测救援分公司,河北沧州062350

出　　处：《现代工程科技》2022年第10期8-11,共4页Modern Engineering Technology

摘　　要：针对机车滚动轴承工作中容易出现故障的现象,提出基于RBF神经网络和小波包的轴承故障诊断方法。对轴承故障中的小波包进行信号的分解和处理,采用降噪处理和特征值处理,然后进行结构重构,引入RBF神经网络对样本进行训练,得出误差在满足要求之内。对3层和4层小波包分解进行分析和试验,最后结合RBF神经网络对滚动轴承进行故障识别,该方法对于轴承故障能够做出准确判断。Aiming at the phenomenon that the locomotive rolling bearing is prone to failure,a bearing fault diagnosis based on RBF neural network and wavelet packet is proposed.The wavelet packet in the bearing fault is decomposed and processed by noise reduction and eigenvalue processing,and then the structure is reconstructed.The RBF neural network is introduced to train the samples,and the error is within the requirements.The 3-layer and 4-layer wavelet packet decomposition are analyzed and tested.Finally,the RBF neural network is used to identify the fault of the rolling bearing,and the bearing fault can be accurately judged.

关 键 词：机车 故障诊断 RBF神经网络 小波包 训练样本 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]