Feature recognition of state signal of electromechanical integration railway turnout over health parameters using CMOS area array technology

Abstract

Abstract Railway turnout is the most important line connection equipment of high‐speed railway system. In order to improve the monitoring and maintenance of high‐speed railway turnout gap, we avoid the large impact of train operations on the turnout. With the increase of train speed, the impact of train operation on turnout is increasing. Therefore, the requirements for monitoring, maintenance, and adjustment of high‐speed railway turnout gap are also gradually improved. In this paper, the feature recognition research of mechatronic railway turnout state signal based on CMOS area array technology is proposed. First, it introduces the collection of turnout health parameters, the collection and processing of gap data in turnout based on CMOS, and the image feature recognition algorithm of adaptive threshold. The problem of inconvenient collecting the bifurcation image showing the gap is solved, and the pollution affects the accuracy. The experimental results show that, according to the fault mechanism of turnout system, turnout faults are mainly divided into electrical faults and mechanical faults. On‐site maintenance: the records show that the mechanical failure accounts for more than 90% of turnout failures. Fault category is the database of fault diagnosis model training. According to the structural principle and fault record description of turnout, there are 12 categories of turnout faults. The feature recognition of mechatronic railway turnout state signal based on CMOS area array technology proposed in this paper has certain validity and accuracy.