Multiple-model Based fault diagnosis for actuators of rail vehicle suspension

This paper is concerned with the design of a multiple-model based fault diagnosis scheme for suspension system of rail vehicles in order to accomplish the task of detecting and isolating faults with reference to multiple actuators. Considering the disturbance caused by rail track unevenness, the multiple actuator-faulty dynamic models together with corresponding observers are set by means of the proposed design procedure. Simulation results illustrate the effectiveness of fault alarm signals as well as the robustness of disturbance and modeling uncertainty of the railway suspension system.