摘要
在锂萃取实验实现自动化控制基础上,针对锂萃取率目前不能在线测量的问题,分别采用RBF神经网络和小波神经网络对锂萃取率软测量模型展开研究。先从锂萃取实验获取基础实验数据,再把实验数据分为训练和预测数据,分别采用RBF神经网络和小波神经网络,对锂萃取率软测量模型进行了多输入单输出和多输入多输出模型试验。试验表明,小波神经网络比RBF神经网络,具有较好的泛化能力,建立了多输入单输出和多输入多输出锂萃取率软测量模型。
On the basis of the automatic control of extraction of lithium experiment,for the problem that extraction of lithium rate can not be measured online at present,the soft measurement model of extraction of lithium rate was researched based on RBF neural network and wavelet neural network.The basic experimental data were obtained from the extraction of lithium experiment,and the experimental data were divided into training and prediction data.The RBF neural network and the wavelet neural network were used respectively.The multiinput single output and multiinput multioutput model were tested for the soft measurement model of extraction of lithium rate.Experiments show that the wavelet neural network has better generalization ability than RBF neural network,and a soft measurement model for the extraction of lithium rate of multiinput single output and multiinput multioutput was established.
作者
何王金
于广平
郭清达
时东
HE Wang-jin;YU Guang-ping;GUO Qqing-da;SHI Dong(Shenyang Institute of Automation in Guangzhou,Chinese Academy of Sciences,Guangzhou Guangdong 511458,China;Qinghai Institute of Salt Lakes,Chinese Academy of Sciences,Xining Qinghai 810008,China;South China University of Technology,Guangzhou Guangdong 511458,China)
出处
《计算机仿真》
北大核心
2021年第2期174-179,共6页
Computer Simulation
关键词
锂萃取
锂萃取率
软测量
神经网络
Extraction of lithium
Extraction of lithium rate
Soft sensing
Neural network