Two-step MPC for systems with input non-linearity and norm-bounded disturbance

作     者：Wang, Jun Ding, Baocang Wang, Yong 

作者机构：Xi An Jiao Tong Univ Sch Elect & Informat Engn Dept Automat Xian 710049 Shaanxi Peoples R China 

出 版 物：《IET CONTROL THEORY AND APPLICATIONS》 (IET控制论与应用)

年 卷 期：2019年第13卷第2期

页      面：183-190页

核心收录：

学科分类：0808[工学-电气工程] 08[工学] 0804[工学-仪器科学与技术] 0811[工学-控制科学与工程] 

基　　金：National Key R&D Program of China [2017YFA0700300] National Science Foundation of China NSFC-Zhejiang Joint Fund for the Integration of Industrialization and Informatization [U1509209] 

主　　题：predictive control linear algebra stability nonlinear control systems iterative methods closed loop systems Riccati equations control nonlinearities linear systems input nonlinearity norm-bounded disturbance two-step model predictive control linear part unconstrained MPC problem desaturation Hammerstein systems two-step MPC Riccati iterative equation nonlinear algebraic equation quadratic boundedness technique stability intermediate control law closed-loop system stability quadratic convergence quadratic cost function minimisation sufficient conditions 

摘      要：In this study, a novel two-step model predictive control (MPC) for Hammerstein systems subject to norm-bounded disturbance is addressed. In the first step, the intermediate control law for the linear part of the system is posed as the solution to the unconstrained MPC problem that minimises a quadratic cost function over a given finite time, for which the solution is determined by a novel Riccati iterative equation. In the second step, the actual control move is obtained by solving non-linear algebraic equation group and desaturation. The quadratic boundedness technique is used to specify the stability for closed-loop system with norm-bounded disturbance, and the sufficient conditions for quadratic convergent of the system state are presented. Simulation results demonstrate the effectiveness of the proposed approach to this class of systems.