摘要
对于输入有约束的LPV时滞系统,提出了准M in-M ax鲁棒模型预测控制.将无限时域“最坏情况”的目标函数分解为当前性能指标和终端代价项,给出了新的鲁棒性能指标上界和系统稳定的充分条件,通过求解LM I的凸优化获得状态反馈控制律,而LM I的可行性保证了闭环系统的稳定性.最后通过truck-tra iler系统仿真验证了所提方法的有效性.
In this paper quasi-rain-max robust model predictive control is proposed for the polytopic linear parameter varying （LPV） time-delay systems with constrained control input. The infinite horizon＇worst-case＇ objective function is transformed to the terminal penalty and objective functions at current time. The sufficient conditions of stability and a new upper bound on the robust performance index are presented for time-delay systems. The feedback control law is obtained by convex optimization involving linear matrix inequalities. It is shown that closed loop stability is guaranteed by the feasibility of linear matrix inequalities. The simulation results of truck trailer system verify the effectiveness of the proposed method.
出处
《武汉理工大学学报（交通科学与工程版）》
2006年第3期373-376,共4页
Journal of Wuhan University of Technology（Transportation Science & Engineering）
基金
国家自然科学基金重点项目资助(批准号:50232030)
关键词
LPV时滞系统
准Min—Max预测控制
LMI
鲁棒
稳定性
linear parameter varying time-delay systems
quasi-rain-max predictive control
linear matrix inequalities
robust
stability
