Active Control Algorithms Of Lateral And Roll Attitude For High Speed Trains

With the rapid development of High-Speed Train (HST) program in China, the requirements of safety operation and ride comfort of HST has improved. As the vehicle’s dynamic performance is greatly affected by the track irregularity, vehicle speed, wind disturbance and the aerodynamic effect, the attitude adjustment and vibration control problem is becoming more and more important yet challenging to address. The vehicle’s suspension system has the function of support, guidance, vibration isolation and reduction, so it can abate the deterioration of its dynamic performance, and maintain the safety of train operation. However, improvement possibilities by means of passive suspension technology eventually reaches a limit. Therefore, active suspension technology is considered as an alternative solution for this issue, since it offers better possibilities of improving the vehicle’s dynamic performance compared to the passive solution. This thesis investigates the active lateral and roll attitude control problem for a high-speed railway vehicle, and some researches have been done as follows:(1) A more suitable and practical dynamic model with unknown time-varying suspension parameters, unexpected disturbances, as well as track irregularities is established. The simulation of this vehicle model running on linear track and curved track without any control action is conducted respectively. Based on the simulation result, some running characteristics of the original vehicle mode are summarized.(2) Two kinds of neuro-network based adaptive control schemes are developed, one of which relies on the system gain matrix, and the other is not. Both of the two control schemes have good robustness against the modeling uncertainties, the unknown time-varying of the system parameters and the uncertain surrounding environment. The difficulty engendered by the non-symmetric positive definite gain matrix is overcomed during the design of the control scheme which does not rely on system gain matrix. This kind of control scheme has simple structure and a smaller count of computations. Rigorous proof of the stability of the proposed control methods are provided based on Lyapunov stability theory.(3)An active fault tolerant control strategy is designed for the system with the incomplete actuator faults. This makes the control strategy more practical, and can be used in similar system with the same model characteristics. The closed loop stability of the control strategy is proved through Lyapunov theory and numerical simulation with MATLAB is conducted to verify the effectiveness of the controller.