Analysis Of Model Predictive Control Characteristics Under The Influence Of Feed Forward Variables

In the field of industrial control,especially in the field of chemical control,the common control algorithm is difficult to control because of the coupling effect between the variables in the control process.To meet the control effect,in the process of industrial control input and output often exist constraints,therefore has the decoupling performance and strong robustness and can effectively deal with input output constraint predictive control are widely used.The feasible region of the predictive control is a very important parameter for the feasibility and stability of the reaction control.The variation of the feasible region can directly affect the control effect of the system.Therefore,it is very important to study the influence of each variable on the system from the perspective of feasible region.In the actual production process,there are a lot of disturbances,because of the strong robustness of predictive control,partial disturbance can be effectively controlled,so that the control effect of the system is not affected.But for the existence of the system can be measured disturbance,because of its change law is known,the impact on the system is predictable.Current scholars have not studied the impact of this part of the disturbance to the system.This part of the interference is introduced into the system as a feed forward variable,and the addition of these variables is bound to affect the system's feasible region.If we do not give priority to this part of the feed forward variables to the system,on the basis of the control algorithm,constraint optimization and so on can not reflect the real situation of the system.In view of this problem,this paper presents the influence of feed forward variables on the feasible region of predictive control.By analyzing the influence of feed forward variables on the feasible region of the system,we can get the effect of feed forward variables on the control effect of the system.In the predictive control,the feed forward variable is part of the system itself,and the other part is made up of the actual production.This part of feed forward variables are transformed from the operation variables: because of the need of the actual production,the control function of the part of the control variables is released.This portion of the feed forward variable may be re converted back to the operating variable.For the state space predictive control,variable conversion will bring about the change of the system structure.System in the "square system" and "non party system" between the conversion process will give the system a significant impact on the feasible region,thusaffecting the control effect of the system.In this paper,based on the study of the influence of feed forward variables on the control effect of the system,this paper further explores the influence mechanism of variable transformation on the feasible region of the system.The mapping idea runs through this paper,and the research on the influence of the variables on the feasible region can be expressed directly by the geometric form.