Track switch fault diagnosis method and system based on support vector machine

Abstract

The invention provides a track switch fault diagnosis method and system based on a support vector machine. The method comprises the steps of: obtaining training samples, wherein the training samples include historical fault current power curves of track switch actions corresponding to a plurality of kinds of track switch fault types; according to factors such as expert experience and fault mechanisms, extracting morphological characteristics of the training samples; selecting a core function of a support vector machine classifier, and determining the parameters of the core function according to the morphological characteristics of the training samples so as to train the support vector machine classifier; collecting current power curves of the track switch actions; and classifying the current power curves according to the support vector machine classifier, and determining whether the track switch is faulted and the fault type of the track switch according to the classifying result. By adopting the method, personnel monitoring and maintenance cost of the track switch is effectively lowered, whether the track switch is faulted and the fault type of the track switch can be automatically and accurately determined, the monitoring of the track switch is convenient, and the maintenance of the track switch is convenient.

Description

technical field [0001] The invention relates to the technical field of high-speed train signal systems and fault diagnosis, in particular to a method and system for fault diagnosis of a turnout based on a support vector machine. Background technique [0002] In recent years, with the rapid development of high-speed rail, there is an urgent need for the steady improvement of high-speed rail equipment fault diagnosis and operation and maintenance technology, especially for the maintenance of the important signal equipment of high-speed rail-turnouts, which are used to make locomotives and vehicles transfer from one track to another. Another line connection device for the stock track greatly improves the ability of the train to pass the line. Because the turnout has the characteristics of large number, complex structure, short service life, limited train speed, low driving safety, and large maintenance and repair investment, it is called the three weak links of the track togeth...

AI Technical Summary

Problems solved by technology

Because the turnout has the characteristics of large number, complex structure, short service life, limited train speed, low driving safety, and large maintenance and repair investment, it is called the three weak links of the track together with the curve and the joint. It plays a vital role in improving the safety of high-speed rail and reducing the maintenance cost of high-speed rail equipment

[0003] The traditional maintenance method for the turnout is mainly based on the experience summary of the on-site staff. The staff will compare the operating current and power curve of the turnout monitored by the microcomputer monitoring system with the current and power curve obtained by the summary, so as to determine the working state of the turnout. However, this method largely relies on the experience of on-site personnel. For inexperienced maintainers, it is very likely that faults will be missed or misreported due to human reasons, which will lead to huge safety hazards. The number of curves is huge, and this monitoring and maintenance method also requires a lot of manpower

Images