基于核偏最小二乘的砷盐净化除钴过程钴离子浓度软测量

来源期刊:中南大学学报(自然科学版)2015年第1期

论文作者:王雅琳 黄凯华 伍铁斌 谢文平 阳春华

文章页码:141 - 149

关键词:砷盐净化除钴过程;核偏最小二乘;软测量;模型在线校正

Key words:cobalt removal purification process with arsenic salt; kernel partial least squares; soft sensor; model online correction

摘    要:针对锌湿法冶炼砷盐净化除钴过程中钴离子浓度无法在线检测给生产优化控制带来困难的问题,建立基于机理模型和核偏最小二乘(kernel partial least squares, KPLS)参数辨识的钴离子浓度软测量模型;考虑到过程具有时变性,根据所建立的软测量模型特点,提出一种双向递归KPLS模型参数更新和滤波修正相结合的模型在线校正方法,以提高软测量模型精度;同时,采用基于主元分析和贝叶斯分类的异常值在线检测方法实现对参数辨识相关检测量的实时异常值在线检测,保证用于参数更新数据的有效性。研究结果表明:所建钴离子浓度软测量模型跟踪效果好,满足实际生产过程预测精度要求,解决了钴离子浓度无法在线检测给优化控制带来的困难,可为生产过程的优化控制提供指导。

Abstract: Considering that the cobalt ion concentration in the cobalt removal purification process with arsenic salt can not be measured online, which makes it difficult to control and optimize the production process, a soft sensor model based on the mechanism and kernel partial least squares was built to predict cobalt ion concentration online. Considering the time-varying characteristics of the process, an online correction method was proposed to improve the precision of the soft sensor model by combining the bidirectional recursive KPLS model parameters update with the filtering correction. Meanwhile, in order to ensure the effectiveness of data samples which were used to model update, an online outlier detection method based on principal component analysis and Bayesian classification was adopted. The results show that the soft sensor model for cobalt ion concentration has good tracking performance. It meets the detection requirement for on-site process and solves the problem of optimal control, so it can provide effective operation guidance for control and optimization of the industrial process.

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号