Track Circuit Fault Diagnosis based on APSO-GMM

Abstract

In view of the huge track circuit system and various faults, this paper proposes a Gaussian mixture model of particle swarm optimization algorithm with adaptive inertia weight to diagnose various fault modes of track circuit. EM algorithm in Gaussian mixture model is easy to be interfered by initial value and fall into local optimum, which leads to unstable diagnosis results and low diagnostic accuracy. The adaptive inertia weight particle swarm optimization algorithm is used to improve the Gaussian mixture model, find the optimal initial value for the model, and improve the stability and fault diagnosis ability of the model. The experimental results show that the improved model is more stable and accurate than the original model or the improved model using single particle swarm optimization algorithm.