Robust Model Predictive Control For Industrial Processes With Time-Varying Delay And Actuator Saturation

Due to the complex characteristics of time-varying delay,actuator saturation,unknown disturbance and nonlinearity,the stable control and safe operation of the production device are difficult.The traditional control method has been unable to meet the process requirements,an effective control method is needed to deal with the production equipment with complex characteristics.Based on previous research work,aiming at the problems existing in industrial production equipment,this thesis studies the new robust predictive control method of industrial process by combining robust predictive control,linear matrix inequality,H₂/_H∞control theory,Lyapunov stability theory and fuzzy control theory.The main tasks are as follows:（1）A robust model predictive control method with mixedH₂/_H∞index is studied for a class of complex industrial control systems with actuator saturation,time-varying state delay and bounded disturbance.Firstly,the convex combination method is used to deal with the saturation term,which greatly reduces the complexity of dealing with actuator saturation characteristics.The Lyapunov-Krasovskii functional depending on the upper and lower bounds of time-varying delay is constructed.Secondly,the disturbance rejection problem of mixedH₂/_H∞robust model predictive control in the sense of norm boundedness is studied.TheH_∞index is regarded as the constraint to be satisfied and theH₂ index is taken as the optimized performance index.The complex optimization problem is transformed into a convex optimization problem of linear matrix inequality,which greatly reduces the difficulty of solution and the amount of calculation.The recursive feasibility and robust stability of the robust predictive control optimization problem are proved.Finally,the simulation research is carried out in the continuous stirring reactor and the water tank level control system.The results show that the proposed method has good practicability and engineering value.（2）A robust fuzzy predictive control method based on T-S model is proposed for a class of uncertain systems with time-varying delay,actuator saturation,nonlinearity,and unknown disturbance.The T-S fuzzy model is established by using several linear sub-models and nonlinear membership functions,and the linear state space model is obtained by weighted linearization of local sub-models.Based on this model,a robust fuzzy predictive control law is further designed,and a Lyapunov-Krasovskii functional depending on the upper and lower bounds of time-varying delay is constructed to make the time-varying delay nonlinear system with actuator saturation and unknown disturbance asymptotically stable.The sufficient conditions for stability are obtained by solving linear matrix inequalities.The simulation results of continuous stirred tank reactor show that the proposed method has strong robustness and good application value.