Application of Weighted Fuzzy Reasoning Spiking Neural P Systems to Fault Diagnosis in Traction Power Supply Systems of High-speed Railways

This paper discusses the application of weighted fuzzy reasoning spiking neural P systems (WFRSN P systems) to fault diagnosis in traction power supply systems (TPSSs) of China high-speed railways. Four types of neurons are considered in WFRSN P systems to make them suitable for expressing status information of protective relays and circuit breakers, and a weighted matrix-based reasoning algorithm (WMBRA) is introduced to fulfill the reasoning based on the status information to obtain fault confidence levels of faulty sections. Fault diagnosis production rules in TPSSs and their WFRSN P system models are proposed to show how to use WFRSN P systems to describe different kinds of fault information. Building processes of fault diagnosis models for sections and fault region identification of feeding sections, and parameter setting of the models are described in detail. Case studies including normal power supply and over zone feeding show the effectiveness of the presented method.