Sliding Prediction Algorithm Discrete Nonlinear System Control

With the characteristic of simple implementation and good robustness for externaldisturbance and unmodeled dynamics, sliding mode control can overcome the uncertainty ofmodel. Predictive control, which can change the global optimization for local rollingoptimization, has good control performance for its convenient modeling. In this paper, byintroducing the predictive control strategy into the discrete sliding mode control system, twodiscrete sliding prediction models are achieved which are based on the power exponentialfunction. Then, two explicit sliding mode predictive controllers are designed, which make thechattering phenomenon weaker, and guarantee the robust stability of the system. The mainresults of this paper are as follows:1) An explicit sliding mode predictive control algorithm based on power-functions ispresented for a class of discrete-time nonlinear systems with disturbance. The main idea is: Byemploying a special model to predict future sliding mode value, utilizing power-functionreaching-law to design referenced track and combining predictive control strategy with slidingmode control, an explicit discrete-time sliding mode predictive control law is constructed forthe controlled system. The closed-loop system is proved having advantages of sliding modecontrol and predictive control, and it weakens the chattering phenomenon effectively.2) An explicit higher order sliding mode predictive control algorithm is presented for aclass of nonlinear uncertain systems: By applying higher order sliding mode method andpower exponential reaching law, combining with the predictive control strategy, a higher ordersliding mode prediction model is constructed, and the explicit higher order sliding modepredictive controller, which assurance the method is simple, easy to implement and the strongrobustness, but also effectively weakened the chattering phenomenon of sliding mode controlsystem.