The Turnout Abnormality Diagnosis with Semi-supervised Learning Method

Abstract

The turnout abnormality is very easy to cause the traffic accident or influence the efficiency due to the operating environment of railway transportation in China. However, the existing monitoring means are relatively backward, and the more mature automatic diagnosis method is lacking. In this study, a method based on semi-supervised learning algorithm for abnormal state diagnosis of turnout action curve is proposed. The method is used to analyze and extract the electrical characteristics of the turnout by using the turnout action curve and the static and dynamic properties collected by the railway centralized monitoring system. The support vector machine model is used to construct the initial classifier with a small number of labeled samples, and the labeled samples are expanded from a large number of unlabeled samples. The diagnosis model is constructed by using unlabeled data with a small amount of labeled data, and the switch curve is analyzed and diagnosed. The experimental results show that the method can automatically diagnose turnout electrical characteristics with high accuracy. Compared with supervised learning, the cost is low, but it can achieve higher accuracy and improve the practicability of fault diagnosis of turnout.