输入受限时滞系统的鲁棒模型预测控制  

Robust model predictive control for time-delay systems with input constraints

作　　者：张军[1,2] 谢荣华[3] 计秉玉[3] 王彪[1] 

机构地区：[1]哈尔滨工业大学 [2]大庆油田有限责任公司,黑龙江大庆163412 [3]大庆油田有限责任公司

出　　处：《电机与控制学报》2004年第4期362-365,共4页Electric Machines and Control

摘　　要：对于输入受限的不确定时滞系统,提出鲁棒模型预测控制方法。当预测状态在不变集之外时,采用双模控制结构,在固定的控制律上引入额外变量保证了输入可行性。为保证闭环系统稳定性,基于Lyapunov-Lrasovskii函数构造了时滞系统特定的鲁棒不变椭圆集,并将椭圆集的设计转化为线性矩阵不等式的凸求解问题。该方法很多计算可以离线进行,从而减少了在线计算量。最后通过仿真验证了该方法的有效性。A robust model predictive control method is proposed for the polytopic uncertain time-delay systems with input constraints. When predictive state is out of invariant set, using dual mode control structure, the extra variable on the fixed state-feedback law is introduced for feasible input. The stability of the closed-loop system is guaranteed by constructing specific robust invariant ellipsoid based on Lyapunov-Krasovskii function, and the design of ellipsoid set is transformed into a convex problem of Lnear Matrix Inequalities (LMIs). Most part of the computation can be performed off-line, and therefore on-line computational burden is reduced. The simulation results verify the effectiveness of the proposed method.

关 键 词：不确定时滞系统 鲁棒模型预测控制 鲁棒不变椭圆集 LYAPUNOV-KRASOVSKII函数 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程] O231[自动化与计算机技术—控制科学与工程]