Design and Health Evaluation of Monitoring Test Platform for Metro Traction Power Supply System

Abstract

Subway traction power supply system is the energy source of subway operation. The normal operation of subway traction power supply system is directly related to the stable operation and safety of subway. If the subway traction power supply system breaks down, it not only will cause economic losses, but also may cause social disorder. Therefore, it is of great significance to study the traction power supply system of subway to ensure the safe operation of subway. The establishment of a reasonable monitoring and testing platform, the simulation analysis of the traction power supply system, and the regular health assessment can provide a strong guarantee for the safe operation of the subway. This paper aims to design a practical test platform through MATLAB modeling, so as to establish a simulation model of subway traction power supply system, and comprehensively monitor and grasp the health status of subway traction power supply system. This paper quantitatively evaluates the health status of subway traction power supply system, applies the principle of multilayer immunity, establishes the traction power supply system of analytic hierarchy process (AHP), and constructs the immune monitoring system of each layer. This method can provide the health status information of subway traction power supply system, subsystems, and components comprehensively, accurately, and intuitively and determine the subsystems and components that need to be maintained.

Review of parametric fault prediction methods for power electronic circuits

“…Generally, multidimension input vector corresponds to one-dimension output (CHI), so as to achieve data dimension reduction or data fusion. In reference [47], the output voltage, peak to peak ripple and output power characteristics of the phase-shifting full bridge circuit are collected and preprocessed as the input of SOM network and the circuitlevel CHI is obtained through training and testing SOM network. It is necessary to point out that some neurons cannot win in the network training, and finally become the dead neurons.…”