准最优稳定的约束非线性系统优先级多目标模型预测控制

Prioritized multi-objective model predictive control of constrained nonlinear systems with quasi-optimal stability

中文关键词  模型预测控制  非线性系统  多目标控制  稳定性  经济优化

英文关键词  model predictive control  nonlinear systems  multi-objective control  stability  economic optimization

中文摘要

      针对具有控制约束的非线性系统,提出一种准最优稳定的优先级多目标模型预测控制策略.采用多目标字 典序方法,根据决策者偏好排序多目标函数,将冲突多目标控制问题转化为具有优先级顺序的单目标序列优化控制 问题.再引入准稳态最优性准则,得到各个优化控制问题目标函数的终端性质,进而建立优先级多目标滚动时域优 化控制问题的递推可行性和闭环性能的近似最优性.数值仿真结果验证了本文策略的优越性.

英文摘要

      This paper proposes a prioritized multi-objective model predictive control (MO-MPC) strategy with quasi optimal stability for nonlinear systems subject to control constraints. Using the multi-objective lexicographical method, the multiple objective functions are sorted based on decision-maker’s preferences. Then the conflict multi-objective opti mization control problem is transformed into a sequence of single objective optimization control problems with the priority order. By introducing the quasi steady-state optimality criterion, the terminal properties of the objective function for each optimization control problem are obtained. Moreover, the recursive feasibility and approximate optimality of closed-loop performance for the prioritized multi-objective receding horizon optimal control problem are established. Numerical simu lation results verify the effectiveness of the proposed strategy.