Research On Discrete Time Output Feedback Sliding Mode Predictive Control

With the development of industrial technology and the further expansion of production scale,the control problem in process industry is becoming more and more complex.On the one hand,the control object of industrial production process not only always has modeling error in system modeling,but also is often affected by external interference.Therefore,the robustness of control algorithm against interference is very important to ensure the quality of production products.On the other hand,in order to obtain better economic benefits,the control algorithm is also required to have the optimal performance.It is difficult to achieve such a control goal by using a single control algorithm.Therefore,this paper studies a composite control algorithm sliding mode predictive control algorithm.Sliding mode predictive control has the advantages of fast convergence speed,strong robustness,simple and convenient design,online rolling optimization of predictive control and optimal performance.It can better meet the actual needs of process industry.In this paper,for discrete-time systems,the theoretical analysis and MATLAB simulation are both used to study the two different combination forms of sliding mode control and model predictive control.For discrete-time systems with unknown disturbances,the output feedback sliding mode predictive control based on differential evolution algorithm is designed.The state observer is used to reconstruct the system state,the model predictive control is used to predict the sliding mode switching function,and the differential evolution algorithm is introduced in the rolling optimization stage of model predictive control to obtain the optimal control law more quickly and accurately.The control algorithm not only weakens the chattering of sliding mode control,but also obtains the optimal control performance.For discrete-time systems with matched disturbance,an integral sliding mode predictive control algorithm is designed.In the integral sliding mode predictive control algorithm,the combination of sliding mode control and model predictive control is different from the general sliding mode predictive control.Under the special structure of the integral sliding mode predictive control,the model predictive control is used to deal with the nominal system,and the integral sliding mode control is used to deal with the external disturbance.So that,the system performance has been greatly improved.For discrete-time systems with unmatched disturbances,an output feedback integral sliding mode predictive control algorithm based on extended state observer is designed.The extended state observer can reconstruct the system state and estimate the external disturbance at the same time,which simplifies the design work and solves the problem of unmeasurable state and external disturbance.The introduction of model predictive control improves large overshoot in integral sliding mode control.