Rail transit fault diagnosis method based on SVM

Abstract

The invention relates to a rail transit fault diagnosis method based on an SVM. The method comprises the following steps: collecting historical monitoring data and real-time monitoring data of rail transit; preprocessing and feature selection are carried out on the monitoring data, during feature selection, according to understanding of the problem and features of the data, partial data related to the problem are selected through experience or a feature selection algorithm, and the partial data are extracted from the original data; vectorizing the feature data; performing model training on the historical monitoring data to generate a corresponding problem classification model; and calculating, analyzing and classifying the real-time monitoring data by the classification model, judging whether a fault exists or not, and obtaining the cause of the fault. According to the invention, an automatic monitoring means is used for replacing manual fault judgment and analysis in massive monitoring signals, so that the labor cost and the fault reason analysis time can be greatly saved, and time guarantee is provided for subsequent maintenance, rescue and other work.

Description

[0001] The present invention is a divisional application of the invention application with the application number 201410009600.4, the application date is January 9, 2014, and the invention title is "A Method and System for Rail Transit Fault Diagnosis Based on SVM". technical field [0002] The invention provides a rail transit fault diagnosis method based on SVM, which relates to technical fields such as railway signal data, railway communication data, railway knowledge data, system alarm data, machine learning, SVM (Support Vector Machine), etc. Data analysis issues for surveillance data. Background technique [0003] At present, there are three main types of monitoring and maintenance products in the field of rail transit (state-owned railways, enterprise railways and urban rail transit): CSM (Centralized Signal Monitoring System), various equipment maintenance machines, and communication network management systems. In order to improve the modern maintenance level of my c...

AI Technical Summary

Problems solved by technology

At present, most of them rely on manual judgment and analysis of faults in massive monitoring data, which requires a lot of labor costs and time for fault cause analysis, making it difficult to provide time guarantee for subsequent maintenance and rescue work. Therefore, it is necessary to study more efficient methods. Data Analysis and Fault Analysis Method for Rail Transit Monitoring

Images