Predictive Control Of Rare Earth Extraction Process Based On Optimized Settings

Because rare earth elements have superior physical,chemical,and electrical properties,they are widely used in military,metallurgical,petroleum,and agricultural industries,and have promoted technological progress in related fields.The extraction and separation of rare earth elements is a long process with non-linearity,strong coupling,serious lag,and complex and variable working conditions.A simple mechanism model cannot describe it,and it is difficult to design a simple and efficient controller to control the process.At this stage,the degree of automation of the rare-earth extraction and separation industrial process is still relatively low,and online detection at the actual site has not been achieved,and experienced operators need to control the extraction process based on their own experience.Therefore,many researchers in the rare earth industry are still conducting continuous research on this.This article starts from the operating cost and economic benefits of the rare earth extraction process.On the premise of comprehensively considering the economic performance indicators and control target requirements of the extraction process,a data-driven modeling idea was adopted to design a method that combines the optimization setting strategy with the model predictive controller.To ensure the stability of the export product quality during the extraction process,the specific research content is as follows:1.In view of the non-linear and strong coupling characteristics of the rare earth extraction process,it is difficult to establish an accurate extraction process model.An Elman neural network modeling method for rare earth extraction process is proposed.According to the process parameters of the extraction site,the content distribution of the components at various levels was analyzed and combined with field data to establish an Elman neural network model for the Ce Pr / Nd extraction process are the prerequisites for appropriate control strategies for subsequent research.2.In order to improve the economic benefits of industrial process,the idea of optimal setting control is constantly improved and developed.Based on the establishment of a high-precision Elman neural network model of the rare earth extraction process,the consumption of fluxes such as detergents and extractants is used as the economic performance index of the control system of the rare earth extraction process.The goal is to maximize the economic benefits of the extraction process.The optimal operating point of the system is obtained by optimizing the set value,and based on those,the design of the predictive controller for the rare earth extraction process is performed.Finally,the validity of the method was verified based on the Ce Pr / Nd extraction process data.3.Aiming at the problems of the complex working conditions of the rare earth extraction process system and the low level of automated production,in order to further improve the economic benefits of complex industrial systems,based on the Elman neural network model,the feedback characteristics of PID control and the prediction of generalized predictive control are combined,a PI generalized predictive control method for rare earth extraction is proposed.The simulation experiments based on the data collected during the Ce Pr / Nd extraction process shows that the improved predictive control method can effectively improve the performance of the controller,is more suitable for the field of rare earth extraction industry,and has more industrial value.In order to achieve the purpose of improving the control effect of the extraction system,this paper designed a system controller based on the optimized settings of the rare earth extraction process model and a generalized predictive control with PI structure controller by establishing appropriate system models.Both simulations have verified the reliability of the two,which has very important practical significance for improving the economic benefits of the rare earth extraction control system.