The iterative linear matrix inequality suboptimal design of decentralized _∞ control

Abstract

An approach is proposed to design decentralized state feedback _∞ suboptimal controllers for LTI interconnected large scale systems. The parametrization theorem of decentralized robust state feedback controllers is developed in two steps and the design condition for the feedback gain is in the form of matrix inequalities. An iterative solution algorithm based on linear matrix inequality (LMI) techniques is proposed to obtain the decentralized feedback gain. The given examples are taken to show the application and the convergence of the algorithm.

Keywords

decentralized control / state feedback / LMI / iterative algorithm / large scale systems

References

[1] WuQ H, MansourM. Decentralized robust control using _∞ optimization technique. Information and Decision Technologies, 1989, 15(2): 59-76

[2] DateR A, ChowJ H. A parametrization approach to optimal H₂ and _∞ decentralized control problems. Automatica, 1993, 29(2): 457-463

[3] HuS S, FanC H, HuW L, et al.. Decentralized robust control of interconnected large-scale system based on _∞ theory. Acta Automatica Sinica, 1994, 20: 356-359

[4] GuiW H, WuM, ChenN. A suboptimal design of _∞ decentralized state feedback control for interconnected systems. J CENT SOUTH UNIV TECHNOL, 1996, 3(2): 159-165

[5] BoydS, GhaouiL E, FeronE, et al.Linear matrix inequalities in systems and control theory, 1994, Philadelphia, SLAM

[6] DoyleJ C, GloverK, KhargonekarP P, et al.. State-space solutions to standard H₂ and _∞ control problems. IEEE Trans Automat Contr, 1989, AC-34(8): 831-847

[7] Shiau J K, Chow J H. Robust decentralized state feedback control design using an iterative linear matrix inequality algorithm. In IFAC 13th Triennial World Congress, USA, 1996. 203–208

[8] GeromelJ B, PeresP L D. Decentralized control through parameter space optimization. Automatica, 1994, 30(10): 1565-1578

[9] IwasakiT, SkeltonR E. All controllers for the general _∞ control problem: LMI existence conditions and state space formulas. Automatica, 1994, 30(8): 1307-1317

[10] Gahinet N A, Laub A J, Chilali M. LMI control toolbox. The Math Works Inc, 1995