Research On Discrete Sliding Mode Control Based On Dynamic Linear Data Model

For solving the problems of disturbance,time delay,error constrained,actuator saturation,system coupling in the general nonlinear discrete-time systems,this thesis are focus on the general nonlinear discrete-time systems and based on dynamic linearization data model,in which,the control technology of integrating dynamic linearization data model and discrete sliding mode control method is studied.Several new discrete sliding mode control methods based on dynamic linearization data model are proposed by combining dynamic linearization technology,discrete integral terminal sliding mode control,disturbance estimation technology,predictive control,prescribed performance control and discrete extended observer technology.The main innovative work of this thesis is summarized as the following aspects:(1)A data-driven discrete integral terminal sliding mode control algorithm based on disturbance estimation technology is proposed for the compact form dynamic linearization data model of the general SISO nonlinear discrete-time systems with unknown disturbances,and the proposed method is extended to the partial form and full form dynamic linearization data model.In order to further improve the tracking accuracy of the proposed algorithm,combing with predictive control method,a discrete integral terminal sliding mode predictive control strategy based on compact form dynamic linearization is proposed.The stability of the proposed approach is guaranteed by theoretical analysis,and the effectiveness of the proposed method is also illustrated by numerical simulation and the experiment on the two-tank water level control system.(2)For a class of general SISO discrete nonlinear time delay systems,a data-driven discrete integral sliding mode control algorithm is proposed based on compact form dynamic linearization data model.The main innovations of the proposed algorithm are:1)the predictor in discrete time is applied to converts the original system with input delay to an equivalent system without the explicit appearance of time delay.2)the control system is designed based on the equivalent dynamic linearization data model without obvious time lag and the discrete integral sliding mode control method,that is,control system design is performed in a new predicted state rather than in the original state.(3)An anti-windup data-driven integral terminal sliding mode control scheme is proposed.Based on compact form dynamic linearization data model,a dynamic anti-windup compensator is introduced in the design process of the discrete integral terminal sliding mode controller,which solves the problem of actuator saturation in the system trajectory tracking process.In order to show the proposed method and its application more clearly,taking the trajectory tracking problem of wheeled mobile robots with input constraints as an example.The proposed control algorithm mainly includes on-line data-driven model identification algorithm,integral terminal sliding mode control algorithm and dynamic anti-windup algorithm.In addition,the stability analysis of the proposed method is given.(4)A data-driven terminal sliding mode control method with tracking error constraints is proposed for compact form dynamic linearization data model with perturbations.By introducing a prescribed performance function to construct unconstrained equivalent transformed error instead of commonly constrained tracking error.A discrete terminal sliding mode controller is designed to further accelerate the sliding mode motion to the sliding surface,and to ensure that the tracking error converges to the predefined domain.(5)A new data-driven discrete integral sliding mode decoupling control algorithm is presented for a class of discrete nonlinear MIMO systems with unknown external disturbances.The unmodeled dynamics,unknown disturbances and the coupling of the compact form dynamic linearization data model are estimated on-line by the discrete extended state observer.Combined with discrete integral sliding mode control strategy,the discrete integral sliding mode decoupling control of discrete nonlinear MIMO systems with disturbances is realized.In order to further improve the control system performance(including overshoot,transient response and steady-state error).A novel data-driven error-constrained discrete terminal sliding mode decoupling control algorithm is proposed.Based on the decoupling of compact form dynamic linearization data model,the algorithm introduces a prescribed performance function in the design process of sliding mode controller,so that the error of each loop after decoupling can converge to a predefined domain,which improves the control performance after system decoupling.Theoretical analysis and simulation experiments are shown that the data-driven sliding mode decoupling controller with prescribed performance can realize the asymptotic tracking control of discrete nonlinear MIMO systems under the influence of interference and system uncertainty.