Event-Triggered Model-Free Adaptive Compensatory Fault-Tolerant Control for Subway Trains under Denial-of-Service Attacks

This paper proposes a model-free adaptive compensatory fault-tolerant control approach based on radial basis function neural networks (RBFNN) and event triggering for subway trains with speed and traction/braking force constraints under denial-of-service (DoS) attacks obeying Bernoulli distribution and uncertain actuator failures. First, a compact form dynamic linearization method is used to transform the complex subway train system into a linear data model. Second, an event-triggered mechanism is designed to reduce the communication and computational burden. A compensation mechanism is developed to handle DoS attacks, while RBFNN is employed to address the nonlinearity caused by actuator failures. Finally, the effectiveness and robustness of the proposed control algorithm are demonstrated through simulation of the subway train system.