Algorithm Research And Experiment Verification On Economic Model Predictive Control For The Continuous Stirred Tank Reactor

The process industry occupies an irreplaceable position in the national economic construction.As the most widely used equipment in the process industry,the performance of continuous stirred tank reactor(CSTR)is closely related to the industrial benefits.Because the market demand,product price,raw material price and other factors related to production are volatile,and because of the high complexity of CSTR,it is an important issue to combine economic considerations with the actual production of CSTR on the basis of ensuring the performance of CSTR.Economic model predictive control(EMPC)has the advantages of rolling optimization and feedback correction.It has the advantages of good control effect and strong robustness.It can effectively overcome the uncertainty in the process,and can well deal with various constraints in the process of manipulating variables and controlled variables.Therefore,this paper uses economic model predictive control method to control CSTR system.Firstly,In this paper,the mechanism of CSTR is modeled and its mathematical model is obtained.Aiming at the problem of information delay and uncertainty of system modeling in the actual production of CSTR,an asynchronous delay economic model predictive control algorithm is designed by using time-varying strategy,so as to improve the steady-state performance of the system.Secondly,When the time-varying economic scale or update frequency(such as variable energy pricing or product demand change)in the actual production process of CSTR is equivalent to the time scale of process / system dynamics,a dynamic economic cost function is needed.Therefore,an economic model predictive control algorithm with time-varying stage cost and an explicit economic cost function that can adapt to adjustment are designed,Then the optimal performance of the system is guaranteed.Thirdly,Considering that CSTR can not run in the optimal mode for a long time in the actual production activities,a real-time economic model predictive control method is proposed,and a real-time implementation algorithm is designed,which can make the continuous stirred tank run in a continuous transient mode and obtain better economic benefits on the premise of ensuring stable and safe production.Finally,The proposed method is simulated and verified by MATLAB,and verified by the process network experimental platform in the laboratory.The Simulation and experimental results show the effectiveness of the proposed method.