Due to the fact that when the actuator components of the semi-active suspension system of the high speed train are prone to malfunction. In view of a solution of this problem, an observer-based integrated sliding mode fault-tolerant control method has thus been proposed. Firstly, a nonlinear dynamic model is constructed of high-speed trains considering actuator faults, followed by a reduction of the system state equation and a design of a sliding mode observer for an estimation of the actuator fault values in real time, thus obtaining the fault expression. Then, a corresponding integral sliding mode fault-tolerant controller is designed with the displacement deviation of the train as the feedback signal, with the output damping force to maintain suspension stability calculated, followed by an analysis of the stability of the sliding mode fault-tolerant controller with Lyapunov stability theory combined together. Finally, based on a simulation and experimental verification, the tracking accuracy of the designed observer reaches around 93% within 0.5 seconds after the fault occurs. Compared with traditional PID control algorithms, the ICSMC control algorithm can stabilize each state variable to near zero domain faster, thus ensuring a stable posture of the trains in operation.