线性不确定系统多目标鲁棒保代价控制  

Multi-objective robust guaranteed cost control of linear uncertain systems

作　　者：王德进[1] 

机构地区：[1]黑龙江大学电子工程学院,黑龙江哈尔滨150080

出　　处：《控制与决策》2004年第3期327-330,共4页Control and Decision

摘　　要：针对一类线性范数有界不确定系统,讨论其多目标鲁棒保代价控制问题.所设计的状态反馈控制器对所有容许的不确定性,保证闭环系统的稳定性,并在非劣(Pareto最优)的意义下,优化多目标代价函数的上界.通过加权因子的选择来解决不同目标之间的竞争问题.采用线性矩阵不等式方法,将问题转化为一个线性凸优化算法.数值仿真算例说明了所提出方法的可解性和有效性.A multi-objective robust guaranteed cost controller is designed for a class of linear norm-bounded uncertain systems. The designed state feedback controller guarantees the stability of the closed-loop system, and in noninferior (equivalently Pareto optimal) sense, optimizes the upper bound of multi-objective cost function for all admissible uncertainties. By selecting a vector-valued weighting parameter appropriately, the confliction among competing objectives is settled. The problem is reduced to a linear convex optimization algorithm via LMI approach. Numerical simulation example shows the feasibility and effectiveness of the method.

关 键 词：多目标控制 不确定系统 保代价控制 鲁棒控制 线性矩阵不等式 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]