稀土萃取分离过程组分含量区间控制方法  

Component content control with zone control for rare earth extraction process

作　　者：陆荣秀[1,2] 何丽娟[1,2] 杨辉[1,2] 张国庆[1,2] 

机构地区：[1]华东交通大学电气与电子工程学院,江西南昌330013 [2]江西省先进控制与优化重点实验室,江西南昌330013

出　　处：《化工学报》2017年第3期1058-1064,共7页CIESC Journal

基　　金：国家自然科学基金项目(51174091;61364013;61673172);国家重点基础研究发展计划项目前期研究专项(2014CB360502)~~

摘　　要：针对稀土萃取过程出口产品的组分含量可以在一定区间范围浮动的要求,提出了一种基于广义预测控制的稀土萃取过程组分含量区间控制方法。首先基于萃取分离过程数据辨识建立组分含量回声状态神经网络(echo state network,ESN)模型;然后针对稀土萃取过程中不同运行工况,采用改进的广义预测控制算法设计组分含量预测控制器,将系统的输出约束纳入求解控制律的优化问题中,使预测控制针对组分含量输出在不同的区域范围采用不同的控制强度,从而实现区间控制同时保证两端出口产品的纯度,最后基于Ce Pr/Nd(铈镨/钕)萃取过程数据的仿真试验验证了该方法的有效性。To meet the requirement that the export product of component content has zone fluctuation in rare earth extraction, the component content control algorithm with zone control based on generalized predictive control of rare earth extraction process is proposed in this paper. Based on the data of the rare earth extraction process, the model of echo state network（ESN） is built up. According to the different running states of the rare earth extraction process, component content predictive controller is designed by using an improved generalized predictive control algorithm, and the algorithm brings the constraint of output variable into the optimization problem for obtaining the control law. Such a design can take various control to different regions of output and realize the stability as possible as zone control of production purity. Simulation results for the Ce Pr/Nd countercurrent extraction process are presented to show the effectiveness of the proposed control approach.

关 键 词：稀土萃取分离 组分含量区间控制 神经网络 动态建模 模型预测控制 

分 类 号：TQ028.8[化学工程]