东北大学学报（自然科学版）

东北大学学报(自然科学版) ›› 2010, Vol. 31 ›› Issue (11): 1554-1557.DOI: -

• 论著 • 上一篇    下一篇

基于智能计算策略的电解液成分软测量方法

王小刚;李春柏;胡轮;   

  1. 东北大学信息科学与工程学院;江西铜业集团公司贵溪冶炼厂;
  • 收稿日期:2013-06-20 修回日期:2013-06-20 出版日期:2010-11-15 发布日期:2013-06-20
  • 通讯作者: -
  • 作者简介:-
  • 基金资助:
    国家自然科学基金资助项目(60974057)

Study on soft sensoring of electrolyte composition based on intelligent computation

Wang, Xiao-Gang (1); Li, Chun-Bai (1); Hu, Lun (2)   

  1. (1) School of Information Science and Engineering, Northeastern University, Shenyang 110004, China; (2) Guixi Smelter, Jiangxi Copper Group, Guixi 335424, China
  • Received:2013-06-20 Revised:2013-06-20 Online:2010-11-15 Published:2013-06-20
  • Contact: Wang, X.-G.
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

430

摘要： 针对电解铜生产过程电解液成分在线实时测量存在的问题,根据铜电解过程的电化学反应机理和生产过程数据,在改进神经网络隐层节点选择策略和优化训练算法的基础上,提出了电解铜生产过程中电解液成分的软测量方法.仿真分析验证了该方法具有较好的数据拟合和泛化性能,现场运行结果也进一步表明该方法能够根据铜电解生产过程的实际变化及时准确地预测电解液成分

关键词: 电解液成分, 软测量, 智能计算, 粒子群算法, 神经网络

Abstract: To achieve the online real-time compositional analysis of the electrolyte for the production of electrolytic copper, a soft sensoring technique was proposed for the compositional analysis according to the electrochemical reaction mechanism and production process data of electrolytic copper, based on the improved selection strategy of hidden nodes in neural network and optimized training algorithm. Simulation results showed that the soft sensoring technique proposed is good at data fitting and generalizability. Furthermore, the in-situ running results also showed that the technique can predict accurately the electrolyte composition in time in accordance to the actual change in the electrolytic process.

中图分类号: