摘要
目前铁路沿线铺设的ZPW-2000系列轨道电路占很大一部分,由于自然、人为等因素,轨道电路会出现故障,针对轨道电路故障的多样性、复杂性、诊断困难等问题,提出基于PCA和PSO-SVM的ZPW-2000轨道电路智能故障诊断方法。首先,对影响因素进行主成分分析,提取了主要的影响因素,将输入维数降低,然后建立常见故障的支持向量机诊断模型,其次采用PSO算法优化SVM模型参数,最后采用某电务段提供数据进行故障划分和诊断,得到较好的诊断效果。
at present,zpw-2000 series track circuits laid along the railway account for a large part.Due to natural and man-made factors,the track circuit will have faults.In view of the problems of the diversity,complexity and diagnosis difficulties of track circuit faults,an intelligent fault diagnosis method of zpw-2000 track circuit based on PCA and pso-svm was proposed.First,principal component analysis on the influencing factors,and to extract the main influence factors,the input dimension is reduced,and then establish a common fault diagnosis model of support vector machine(SVM),secondly,parameters of the SVM model was optimized by using PSO algorithm,and finally USES a signal depot to provide data for fault classification and diagnosis,get good diagnosis effect.
出处
《数码设计》
2019年第19期35-36,共2页
Peak Data Science
基金
沧州市重点研发计划指导项目(183103001)《基于数据的ZPW-2000A型轨道电路智能故障诊断方法研究》。
