稀土萃取过程的广义预测解耦控制

考虑稀土萃取过程具有多变量、强耦合特性,提出一种基于广义预测解耦控制(GPDC)的稀土萃取过程控制方法。首先针对模型未知的强非线性稀土萃取过程,构建基于极限学习机的组分含量系统模型,并依据模型特点设计多个稀土萃取过程GPDC控制器;然后为降低各控制回路间的耦合性,在控制器的性能指标中引入校正策略,通过回路中模型预测值与参考值的偏差自适应调整偏差权重;最后基于CePr/Nd萃取过程中采集的数据进行GPDC与常规广义预测控制器进行对比仿真实验。仿真结果显示本文采用的GPDC算法能大幅度降低控制量的超调量,控制效果显著,这为解决稀土萃取过程中多变量强耦合的优化控制问题提供了借鉴。

Generalized Predictive Decoupling Control for Rare Earth Extraction Process

Considering the multivariable and strong coupling characteristics of rare earth extraction process,a method of rare earth extraction process based on generalized predictive decoupling control(GPDC)is proposed.Firstly,for the strong nonlinear rare earth extraction process with unknown model,a component content model based on extreme learning machine is constructed,and several GPDC controllers are designed according to the model characteristics.To reduce the coupling between the control loops,a correction strategy is introduced into the performance index of the controller.The deviation weight can be adjusted adaptively by calculating the deviation between the model prediction and the reference value in the control loop.Finally,comparing the GPDC with the conventional generalized predictive controller based on the data collected in CePr/Nd extraction process,the simulation results show that the proposed method can greatly reduce the overshoot of the control variables,and improve the control performance significantly,which provides a reference for solving the problem of multi-variable and strong coupling optimization control in rare earth extraction process.