基于遗传算法优化BP神经网络的稀土萃取过程建模
The Modeling of the Rare-Earth Extractive and Separate Process Based on Genetic Algorithm and BP Neural Network

针对稀土萃取分离过程工艺复杂/难以对其建立精确的过程控制模型,提出了稀土萃取过程遗传算法优化BP神经网络建模方法.根据现场工艺参数,确定稀土萃取过程级数和各进料设定流量; 运用串级萃取理论对稀土萃取过程产生各级组分含量数据分析; 应用Matlab7.10数学软件进行分析计算,为确保2端出口产品最终达到所需纯度对过程进行研究分析,并以CePr/Nd萃取过程为例进行建模研究,对实际生产过程工艺操作控制具有一定的借鉴意义.
To solve the difficulty of rare earth separation and establishing precise process control model of the rare earth extraction process,the genetic algorithm-BP neural network modeling method of the rare earth extraction process is proposed.Based on the process parameters of the site,the stages and the flow-rate of each feed of the rare earth extraction process are determined.According to the countercurrent extraction theory,the component content data at all levels of rare earth extraction process is analyzed.Through the analysis and calculation by using Matlab7.10,the extraction process is studied to ensure that both end of export products eventually can reach the required purity.For the example of the modeling studies on the CePr/Nd extraction process,it shows the certain reference significance to process operation control in the actual production process