| Many industrial processes are subject to input or output limits,for such a constrained system,this paper proposed a constrained data driven optimal iterative learning control(constrained-DDOILC)scheme,the main work and innovation of this paper are as follows:1)We classified the traditional constrained optimal iterative learning control(OILC)method,and the limited range is divided into three aspects: input constraints,output constraints and input and output constraints,at the same time,the control methods used when there are uncertain parameters in the control system are discussed.2)The constrained-DDOILC scheme is proposed for a class of constrained nonlinear discrete-time systems,this scheme does not require any model information,it is designed by the tool of quadratic programming,using I/O data and introducing a new iterative dynamic linearization method when considering the I/O constraints.The proposed scheme is data-driven and without modeling,so the design of unmodeled dynamics problem does not exist in the process,the control examples for the simulation ensured the robustness of the controlled system.In this paper,the control method based on Lifted is put forward,and the simulation is carried out by using some examples.3)In this paper,the control scheme based on Lifted technique is extended to the Non-lifted technique for nonlinear discrete-time systems with input and output constraints,and the corresponding convergence proof and simulation verification are carried out.4)This paper proposes a constrained data-driven optimal point-to-point iterative learning control(constrained-DDOPTPILC)scheme,which only uses the error information of specific points,tracking the desired one or several points of the given system.The scheme greatly reduces the amount of calculation,eases the cost of the industry and improves the efficiency.The simulation results show that the proposed method is more effective than the traditional methods. |
PDF Full Text Request