Adaptive Dynamic Surface Control for Speed Tracking High-Speed Train Systems

In this paper, we study the speed tracking problem of a nonlinear high-speed train system, which involves coupling between the cars and unknown slow time-varying parameters. The four-car example, which is similar to [4], has also been chosen for system analysis and design. Inputs to the model affect the second and third cars, and aerodynamic drag is also considered. An adaptive controller incorporating dynamic surface control (DSC) is developed to handle the nonlinear and unknown parameters present in the system. The proposed control law overcomes the “complexity explosion” problem associated with backstepping applied to high-order train systems and avoids the introduction of unobservable states. We demonstrate that all signals in the control system are semiglobally uniformly bounded. In addition, the virtual control law for maintaining inter-unit stability and the correlation between control gain and filter time constant are provided based on a candidate Lyapunov function. Finally, the simulation results of CRH2 high-speed train are given to verify the performance of the designed adaptive control system.