基于观测器的不确定广义时滞系统鲁棒预测控制  

Robust model predictive control of singular systems with delayed-state and parameter uncertainty based on state observer

作　　者：刘晓华[1] 杨园华[2] 

机构地区：[1]鲁东大学数学与信息学院,山东烟台264025 [2]山东大学控制科学与工程学院,济南250061

出　　处：《控制与决策》2009年第4期606-610,616,共6页Control and Decision

基　　金：国家自然科学基金项目(60774016)

摘　　要：针对一类不确定广义时滞系统,讨论了其基于观测器的鲁棒预测控制问题,给出了系统观测器型预测控制器的设计方法.通过构造带有误差项的Lyapunov函数,应用线性矩阵不等式,将无穷时域二次性能指标"min-max"优化问题转化为凸优化问题,得到了鲁棒预测控制器存在的充分条件和显式表达式.证明了优化问题在初始时刻的可行解能保证广义闭环系统渐近稳定且正则无脉冲.仿真实例验证了所提出方法的有效性.The problem of observer-based robust predictive control for the singular systems with uncertainties and time-delay is presented. The design method of robust predictive observer-based controller is proposed. By constructing the Lyapunov function with the error terms, and with the method of LMIs, the infinite time domain ＂min-max＂ optimization problems are converted into convex programming problems. Then the sufficient conditions for the existence of this control are derived. The robust stability of the closed-loop singular systems is guaranteed by the initial feasible solutions of the optimization problems. The regular and the impulse-free of singular systems are also hold. A simulation example illustrates the effectiveness of this method.

关 键 词：预测控制 广义时滞系统 线性矩阵不等式 状态观测器 不确定性 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]