Robust Predictive Control Of Uncertain Nonlinear Systems With Multiple Time Delays

In many complex industrial projects,there is a new type of computer control algorithm-Model predictive control.Due to its ability to adapt to complex production,it is preferred by the majority of industrial process control field.For most actual industrial systems,it is unavoidable to emerge the nonlinear,time lag and uncertain pheno menon,while the three problems-stability,robustness and online optimization have always existed,therefore,it is worth to study the nonlinear time-delay uncertain system.Predictive control has the characteristics of rolling optimization,which can optimize online repeatedly,and the robust control can deal well with model uncertainty,the combination of them can effectively solve the robust predictive control for nonlinear uncertain systems with timedelay.This paper uses two methods: Lyapunov stability theory and linear matrix inequality,to engage in studying the robust predictive control problem with time-delay parameters uncertain nonlinear system.The main contents of this paper are as follows:1.For a class of single time-delay uncertain system under the nonlinear perturbation,this Chapter discusss its robust predictive control problem and the robust stability problem,based on state feedback,constructs the Lyapunov functions related to time-delay,and transforms the rolling optimization problem of predictive control into solving LMI problem.So as to approximately solute convex optimization problem in infinite horizon,design a piecewise continuous predictive controller,as well as solve the feasibility of online optimization and ensure the robust stability of the closed-loop system.2.On the basis of the previous chapter,Chapter Four considers the multi-time-delay singular system,and discusses the robust predictive control problem based on state feedback.Under the "minimum-maximum" performance indicators,this chapter designs a new predictive control algorithm,solves the repeated online optimization problem,and uses the Lyapunov function associated with time-delay to give out the sufficient conditions for the impulse-free asymptotically stable of the closed-loop system.