Robust Adaptive Iterative Learning Control for Multitrain System With Actuator Saturation and State Constraints
针对高速列车动力学模型不确定和运行过程非严格重复的问题，提出一种鲁棒自适应迭代学习控制方法，通过引入非线性辅助系统补偿执行器饱和，并开发屏障方法实现速度和位置状态约束，确保多列车系统在安全范围内运行。
In this article, the cooperative control issue is investigated for multitrain system with actuator saturation and state constraints. Considering the uncertainty of the high-speed train dynamics model and the nonstrict repetition of the operating process, a robust adaptive iterative learning control (RAILC) method is proposed for multitrain system cooperative control, where a control-based nonlinear auxiliary system is introduced to realize the finite-time compensation of train actuator saturation. Furthermore, a novel barrier RAILC (BRAILC) method is developed to achieve actively speed and position state constraints, which can ensure the multitrain system maintaining operates within the safe range. As the best of the author’s knowledge, this is the first time that the robust control and adaptive iterative learning control (AILC) are applied in a multitrain system. The Lyapunov function and composite energy function (CEF) are used to analyze the system convergence along the time and iteration axis, respectively. In the time axis of each iteration, the speed and position tracking errors of multiple high-speed train are convergent, and the system state is constrained, thus the proposed controller is reliable in engineering. Furthermore, through iterative learning law, the speed and position tracking error will converge to zero along the iteration axis, which gives the system the ability to intelligently utilize historical data. Finally, the effectiveness of the proposed control method is demonstrated through numerical simulations using data from China railway high-speed 380B (CRH380B) train.