摘要
随着通信、存储技术的不断进步,轨道交通已向“车辆大数据”的方向发展,对故障诊断技术提出了更高的要求。给出了一种FFT+DBN+参数寻优的牵引系统电机轴承诊断方法,完成了无监督特征提取与有监督微调网络模型的构建,解决了网络参数设置难的问题,并有效提高故障识别准确度,为电机轴承故障诊断提供了解决方案,具有很强的工程应用价值。
With the continuous progress of communication and storage technologies,urban rail transit has started the development towards the direction of"Vehicle Big Data",which puts forward higher requirements for fault diagnosis technology.A diagnosis methodology of motor bearings in traction system based on FFT+DBN+parameter optimization is proposed,which completes the unsupervised feature extraction and the construction of supervised fine tuning network mode,solves the problems in network parameter setting,improves the accuracy of fault identification.At the same time,the methodology provides a solution to fault diagnosis of motor bearings and therefore has strong engineering application value.
作者
涂小卫
张士强
王明
TU Xiaowei;ZHANG Shiqiang;WANG Ming(Technology Center,Shanghai Shentong Metro Group Co.,Ltd.,201103,Shanghai,China;不详)
出处
《城市轨道交通研究》
北大核心
2020年第1期174-178,195,共6页
Urban Mass Transit
关键词
地铁
车辆
牵引电机轴承
深度学习
故障诊断
metro
vehicle
traction motor bearing
deep learning
fault diagnosis
作者简介
第一作者:涂小卫,工程师。
