基于核偏最小二乘法的湿法冶金萃取过程建模  

Modeling for Solvent Extraction of Copper Based on Kernel Partial Least Squares Method

作　　者：陈溥[1] CHEN Pu(Liuzhou Railway Vocational Technical College,Liuzhou 545616,China)

机构地区：[1]柳州铁道职业技术学院,广西柳州545616

出　　处：《湿法冶金》2021年第3期260-266,共7页Hydrometallurgy of China

摘　　要：针对铜的溶剂萃取过程,提出了一种混合模型。该模型由参数未知(传质速率)的机理模型串联参数未知的数据预测模型组成,采用核偏最小二乘法建立数据预测模型。并通过仿真与BP神经网络模型进行对比,分析混合模型的性能。结果表明,基于核偏最小二乘方法的串联混合模型有较好性能,可用于优化铜的溶剂萃取工艺。A mixed model for solvent extraction of copper is put forward.The model is composed of mechanism model of unknown parameters(mass transfer rate)and data prediction model with unknown parameters.The data prediction model is established by kernel partial least squares method,and the performance of hybrid model is analyzed by comparing simulation with BP neural network.The results show that the mixed model has good performances and can be used to forecast the solvent extraction process of copper.

关 键 词：湿法冶金 铜 溶剂萃取 机理模型 核偏最小二乘 混合建模 

分 类 号：TF803.2[冶金工程—有色金属冶金]