Adaptive Fault-Tolerant Sliding Mode Control for High-Speed Trains With Actuator Faults Under Strong Winds

This article addresses the fault-tolerant control problem for high-speed trains (HSTs) with actuator faults under strong winds. For the healthy system, a non-singular fast terminal sliding mode surface is introduced into the controller, which ensures the system fast converge to the equilibrium point with finite time, and the radial basis function neural network (RBFNN) with the adaptive compensation term of the error is used to approximate the unknown nonlinear disturbances of the train system in strong winds and predict the time delay generated in the train communication network. Through the RBFNN observer established specifically for actuator failures, the real effective factors are identified. Combining with the identified effective factors and predicted time delay, an adaptive fault-tolerant sliding mode control method is established for train systems. With dynamic parameters and position update strategy guided by negative gradient thought, an improved particle swarm optimization algorithm is proposed to optimize the uncertain parameters of control method, which weakens the chattering phenomenon from the sliding mode controller and improves the control accuracy. Simulation results show that the proposed method has better tracking performance and real-time performance compared with other fault-tolerant control methods under various operating conditions, which ensures the running safety of HSTs under different strong winds.