Operator-based Optimal Tracking Control Of Non-linear System

In this thesis, optimal tracking control of non-linear systems with uncertainties is researched. In the actual industrial systems, most of those systems are nonlinear systems, and the completely accurate models are impossible to get. Because of some Instability factor and various faults which have great influence on the tracking control of nonlinear systems. For those systems robust control is usually needed to ensure the robust stability, and then tracking control is researched. There are many robust control methods, but the robust right coprime factorization method based on operator is a novel method which has a good effect on the systems’ robust control under uncertainties. So the method is used to ensure the robust stability of the system, and then based on this research the optimal tracking control.In general, the design of the tracking controller is based on calculating the inverse of the operator, and the inverse is very difficult to obtain for many nonlinear systems. So many approximate inverse methods are used to design system’s tracking controller. However those methods don’t solve the inverse. For avoiding inversion, two methods are proposed to improve the tracking performance, including based on equivalent substitution optimal tracking control and based on Lipschitz norm optimal tracking control, and another tracking operator is consisted by using Bezout identity. The first scheme can achieve the tracking control, but it is a suboptimal solution. The second method is designed to improve the first one. The main content as following:Firstly, Peltier refrigeration technology and the modeling of Peltier refrigeration system are introduced. Operator-based robust controller is studied to ensure the robust stability of the system by using robust right coprime factorization method. Then, a tracking compensator operator based on equivalent substitution optimal tracking control by using particle swarm optimization(PSO) for constrained conditions is designed to satisfy the tracking performance of the system. The simulation results proof the method could achieve the system’s tracking performance, but it is a suboptimal solution and can’t ensure the system’s dynamic characteristics. So it needs to be improved.Then, based on Lipschitz norm optimal tracking control by using PSO is proposed to improve the first tracking controller defect of suboptimal solution. Then the proposed method is confirmed by the numerical simulation and experimental results.In all, this paper mainly researches the optimal tracking control based on the system’s robust stable. Two methods are proposed to improve the tracking performance, including based on equivalent substitution optimal tracking control and based on Lipschitz norm optimal tracking control, and another tracking operator is consisted by using Bezout identity. Then the proposed method is confirmed by the numerical simulation and experimental results.