Research on Railroad Turnout Fault Diagnosis Based on Support Vector Machine

Turnout is an important signal equipment to ensure the safe operation of train traffic, and its reliability is crucial to the safety and efficiency of railroad transportation. The maintenance of turnouts is mainly based on experience and relevant knowledge for periodic maintenance. There are many problems with this method, such as inaccurate fault location, slow fault analysis, and low work efficiency. Therefore, it will affect the efficiency of driving. Taking S700K type speed switch as the research object, the basic principle of switch control circuit and common switch faults are analyzed, and combined with microcomputer monitoring, the support vector machine (SVM) based algorithm is used for fault diagnosis in view of the complexity of fault diagnosis, and compared with two algorithms of neural network (BP) and Bayesian network (NB). The results show that the fault diagnosis is better by using SVM-based algorithm.