东北大学学报（自然科学版）

东北大学学报(自然科学版) ›› 2008, Vol. 29 ›› Issue (2): 174-177.DOI: -

• 论著 • 上一篇    下一篇

基于支持向量机插值的铜电解液成分检测

袁平;李妍;毛志忠;   

  1. 东北大学信息科学与工程学院;东北大学信息科学与工程学院;东北大学信息科学与工程学院 辽宁沈阳110004;东北大学流程工业综合自动化教育部重点实验室;辽宁沈阳110004;辽宁沈阳110004;辽宁沈阳110004;东北大学流程工业综合自动化教育部重点实验室;辽宁沈阳110004
  • 收稿日期:2013-06-22 修回日期:2013-06-22 出版日期:2008-02-15 发布日期:2013-06-22
  • 通讯作者: Yuan, P.
  • 作者简介:-
  • 基金资助:
    国家自然科学基金资助项目(60374003)

Detection based on support vector machine interpolation for copper electrolytic composition

Yuan, Ping (1); Li, Yan (1); Mao, Zhi-Zhong (1)   

  1. (1) School of Information Science and Engineering, Northeastern University, Shenyang 110004, China; (2) Key Laboratory of Integrated Automation of Process Industry, Northeastern University, Shenyang 110004, China
  • Received:2013-06-22 Revised:2013-06-22 Online:2008-02-15 Published:2013-06-22
  • Contact: Yuan, P.
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

512

摘要： 在电解铜生产过程中,由于检测设备的限制,只能通过每天化验一次得到铜离子和酸根离子含量的检测数据.这些有限数据不足以反应铜电解液成分的变化规律,而基于传统插值法的插入值和现场实际值差距又很大.为解决这一问题,提出了与影响因素相关的软测量模型,并实现对数据的有效预测.鉴于支持向量机在建立软测量模型方面的优势,引入支持向量机对线性插值方法进行改进,并设计权重确定方法,定量推导出软测量估计值和传统线性插入值的权重.软测量建模方法既保证了数据在时间上的连续性,又保证了计算的插入值接近实际值.仿真结果表明了其有效性.

关键词: 插值, 支持向量机, 阴极铜, 权重, 软测量

Abstract: In the process of electrolytic copper production the contents of copper ions and acid ions can only be detected once each day through tests, and they arc inadequate to showing the regularity of variation of copper electrolyte composition. It was found that the values for conventional interpolation often differ greatly from the values obtained actually in site. To solve the problem, a soft measurement model is advanced considering the influencing factors on measurement with relevant data forecasted efficiently. The support vector machine is introduced to improve linear interpolation because of its advantages in developing the soft measurement model. And a method is given to determine the weights for the values estimated by the soft measurement and values for conventional linear interpolation. The approach proposed ensures not only the time continuity of data but also the values for interpolation to approximate to actual values. Simulation results showed its effectiveness.

中图分类号: