不确定系统预测控制多面体终端域的迭代优化方法OA

Constructing terminal constraints is a crucial means to ensure the stability and feasibility of predictive control systems,thus optimizing the terminal region plays a pivotal role in enhancing the performance of predictive control syste-ms.Addressing the challenges posed by uncertain systems in predictive control,this paper proposes an optimization method to solve polyhedral terminal regions,terminal control laws,and terminal penalty functions,achieving iterative volume in-crement of polyhedral terminal regions.Initially,the separation of uncertain disturbances in the system and a coordinate transformation are employed to construct polyhedral terminal constraints ensuring asymptotic stability and optimized feasi-bility of the closed-loop system.Subsequently,by leveraging the full-block S-procedure and the Polya relaxation principle,the terminal constraints containing uncertain parameters are equivalently transformed into a deterministically formulated matrix inequality.A solution method based on linear matrix inequalities is then provided.Furthermore,an iterative opti-mization algorithm for polyhedral terminal regions is designed,theoretically ensuring the iterative increase in the volume of the polyhedral terminal region and overcoming the limitations of matrix symmetric constraints.Finally,the feasibility and the effectiveness of the algorithm are validated through simulation results.