摘要
在城市轨道交通车辆受电弓日常检修过程中,大量检修及故障数据未得到合理利用。针对计划检修已不能满足目前受电弓检修要求的问题,提出了一种基于主元分析和概率神经网络结合的故障诊断方法。该方法运用主元分析法对受电弓日常检修中的初始特征参数进行降维,将降维后特征参数输入到概率神经网络模型中进行故障诊断,判定受电弓故障模式。仿真结果表明,该诊断方法耗时短、正确性高。
During the urban rail transit vehicle pantograph routine maintenance process,a large number of overhaul and fault data are not rationally used.Targeting the problem that planned maintenance can no longer meet the current pantograph maintenance requirements,a fault diagnosis method based on principal component analysis and probabilistic neural network is proposed.This method uses the principal component analysis to reduce dimension of the initial feature parameters from pantograph daily maintenance.The feature parameters after dimension reduction are input into the probabilistic neural network model for fault diagnosis,to determine the pantograph failure mode.The simulation results show that this diagnostic method takes less time and has high diagnostic accuracy.
作者
王宇
刘若晨
WANG Yu;LIU Ruochen(Jiangsu University of Technology,213001,Changzhou,China)
出处
《城市轨道交通研究》
北大核心
2021年第1期88-92,共5页
Urban Mass Transit
基金
国家自然科学基金资助项目(51705221)。
关键词
城市轨道交通
车辆
受电弓
故障诊断
主元分析
概率神经网络
urban rail transit
vehicle
pantograph
fault diagnosis
principal component analysis
probabilistic neural network
