Soft-sensor for copper extraction process in cobalt hydrometallurgy based on adaptive hybrid model

In the process of copper extraction in cobalt hydrometallurgy, the copper concentration of raffinate solution needs to be monitored and controlled simultaneously. It is difficult to measure such concentration online by existing instruments and sensors. Soft sensor technique has been became an online supplement measurement for process monitoring and control. In this paper, an adaptive hybrid modeling method for copper extraction process is proposed. The proposed model is composed of simplified first principle model and block-wise recursive PLS model. The former based on material balancing calculation with some assumptions is used to describe the extraction process in general; and the latter is constructed to compensate the unmodeled characteristic and deal with the time-variant feature. A model rectification strategy is also employed to correct the final output and increase the prediction accuracy. The proposed model has been used in a cobalt hydrometallurgy pilot plant, and the prediction results indicate that the adaptive hybrid model is more precise and efficient than the other conventional models.

[1]  Efstratios N. Pistikopoulos,et al.  A Hybrid Modelling Approach for Separation Systems Involving Distillation , 1999 .

[2]  S. Mishra,et al.  Separation of Co, Ni and Cu by solvent extraction using di-(2-ethylhexyl) phosphonic acid, PC 88A , 1998 .

[3]  B. Kowalski,et al.  Partial least-squares regression: a tutorial , 1986 .

[4]  Jie Zhang,et al.  A recursive nonlinear PLS algorithm for adaptive nonlinear process modeling , 2005 .

[5]  M. Anitha,et al.  Artificial neural network simulation of rare earths solvent extraction equilibrium data , 2008 .

[6]  Ping Wu,et al.  Online dual updating with recursive PLS model and its application in predicting crystal size of purified terephthalic acid (PTA) process , 2006 .

[7]  Sirkka-Liisa Jämsä-Jounela,et al.  Dynamic modelling of an industrial copper solvent extraction process , 2005 .

[8]  Sten Bay Jørgensen,et al.  A systematic approach for soft sensor development , 2007, Comput. Chem. Eng..

[9]  Chris Aldrich,et al.  Modelling of rare earth solvent extraction with artificial neural nets , 1996 .

[10]  K. Helland,et al.  Recursive algorithm for partial least squares regression , 1992 .

[11]  S. Qin.  Recursive PLS algorithms for adaptive data modeling , 1998 .

[12]  S. Wold,et al.  PLS-regression: a basic tool of chemometrics , 2001 .

[13]  Wenjun Jia,et al.  A Hybrid Intelligent Soft-Sensor Method for the Rare Earth Cascade Extraction Process , 2006, 2006 6th World Congress on Intelligent Control and Automation.

[14]  Peter Andersson,et al.  Adaptive Forgetting in Recursive Identification through Multiple Models , 1985 .

[15]  Yongfeng Fu,et al.  MIMO Soft-sensor Model of Nutrient Content for Compound Fertilizer Based on Hybrid Modeling Technique 1 1 Supported by the National Natural Science Foundation of China (No.60421002) and the New Century 151 Talent Project of Zhejiang Province. , 2007 .

[16]  Wenjun Jia,et al.  Multiple Bilinear Models Based Soft-Sensor for Rare Earth Cascade Extraction Processes , 2006, Proceedings of the 45th IEEE Conference on Decision and Control.

[17]  Bhupinder S. Dayal,et al.  Recursive exponentially weighted PLS and its applications to adaptive control and prediction , 1997 .

[18]  Hans-Jörg Bart,et al.  Separation of Cobalt and Nickel by Reactive Extraction – Modeling of Equilibria , 2006 .

[19]  Jana Wichterlová,et al.  Dynamic behaviour of the mixer–settler cascade. Extractive separation of the rare earths , 1999 .

[20]  Daniel Hodouin,et al.  Simulation of a SX–EW pilot plant , 2000 .

[21]  Yi-Zeng Liang,et al.  Monte Carlo cross validation , 2001 .