Research On Optimal Control Of Switched Systems Based On Adaptive Dynamic Programming

As a very important class of hybrid dynamic systems,switched systems have attracted great attention because of their wide application.In recent years,many researchers have done many researches on switched system,and achieved a lot of important results related to the optimization,stability and controllability of the switched system.For the optimal control of switched systems,this paper mainly focuses on the switched systems under the execution saturation constraints and the switched systems with state delay under the execution saturation constraints,and applies the piecewise iterative algorithm and the intelligent optimization method to design the optimal control law.Firstly,the optimal control problem of switched system with state delay is studied under the constraint of execution saturation.For the system with uncertain switching time and unknown switching sequence,an adaptive dynamic programming optimal control algorithm based on the least square method is proposed,and the convergence conditions of executing network weights in the algorithm are analyzed and proved.Finally,an example is given to verify the effectiveness of the algorithm.Secondly,particle swarm optimization(PSO)is applied to the optimal control of the switched system with executive saturation constraints.Firstly,the basic PSO is studied,and the advantages of the basic PSO in optimizing the weights are discussed.In order to improve the performance of optimization,adaptive inertia parameter factor and cross genetic are used to improve the optimization process.The results show that compared with the adaptive dynamic programming algorithm based on the least square,this method has better optimization control effect.Finally,the effectiveness of the algorithm is verified by a practical example.Finally,the main work of this paper is summarized,and the shortcomings in the process of optimal control of switched systems are pointed out,and the future research work is prospected.