Optimal Design Of Anti-windup System

As we all know,actuator saturation exists widely in various industrial control systems and is a common nonlinearity.When the actuator saturates,the performance of the closed-loop systems designed without considering actuator saturation may deterioration and even instability.Thus,the research on the control system subject to saturation has an important theoretical and practical significance.Based on the general anti-windup framework,this paper studies the optimal design of anti-windup compensator for linear time-invariant system.For single-input singleoutput systems,an improved circle criterion is proposed,which reduces the conservatism of anti-windup design significantly;Considering the uncertainty of the object model parameters and bandwidth constraints,applying the improved circle criterion,the sensitivity and complement sensitivity are used as the optimization objective function,so that the optimization design of the anti-windup compensator is converted into a minimax optimization problem,and the optimal parameters of anti-windup compensator are obtained by fminimax function in Matlab.For multi-input multi-output system,the Weston-Postlethwaite’s generalized anti-windup compensation framework is equivalently converted into a parallel connection of a nominal linear feedback system and a dead-zone nonlinear loop,so the optimal design of the compensator is converted to the problem of minimizing nonlinear loop gain;Based on the state space description,the LMI optimization toolbox of MATLAB is used to solve the minimization problem,so as to obtain the optimal parameters of the compensator,which provides a recommended solution for the design of the multivariable anti-windup compensator.Simulation examples verify the effectiveness of the proposed method.