Modified recursive partial least squares algorithm with application to modeling parameters of ball mill load

Recursive partial least squares (RPLS) regression is effectively used in process monitoring and modeling to deal with the stronger collinearity of the process variables and slow time-varying property of industrial processes. Aim at the RPLS cannot solve the modeling speed and the accuracy problems effectively, a modified sample-wise RPLS algorithm is proposed in this paper. It updates the PLS model according to the process status. We use the approximate linear dependence (ALD) condition to check each new sample. The model is reconstructed recursively such that the new samples satisfy the ALD condition. Experimental study on modeling parameters of ball mill load shows that the proposed modified RPLS algorithm is computationally faster, and the modeling accuracy is higher than conventional RPLS for the time-varying process.

[1]  Jian Tang,et al.  Experimental analysis of wet mill load based on vibration signals of laboratory-scale ball mill shell , 2010 .

[2]  Gary Montague,et al.  Soft-sensors for process estimation and inferential control , 1991 .

[3]  Christopher M. Bishop,et al.  Neural networks for pattern recognition , 1995 .

[4]  Shie Mannor,et al.  The kernel recursive least-squares algorithm , 2004, IEEE Transactions on Signal Processing.

[5]  I. Jolliffe.  Principal Component Analysis , 2002 .

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

[7]  Weihua Li,et al.  Recursive PCA for Adaptive Process Monitoring , 1999 .

[8]  A. Höskuldsson.  PLS regression methods , 1988 .

[9]  Wen Yu,et al.  Fuzzy modelling via on-line support vector machines , 2010, Int. J. Syst. Sci..

[10]  Xiaoou Li,et al.  On-line fuzzy modeling via clustering and support vector machines , 2008, Inf. Sci..

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

[12]  C. V. R. Murty,et al.  Experimental analysis of charge dynamics in tumbling mills by vibration signature technique , 2007 .

[13]  Binglin Zhong,et al.  Investigation on measuring the fill level of an industrial ball mill based on the vibration characteristics of the mill shell , 2009 .

[14]  K. Gugel,et al.  Improving ball mill control with modern tools based on digital signal processing (DSP) technology , 2003, Cement Industry Technical Conference, 2003. Conference Record. IEEE-IAS/PCA 2003.

[15]  Xiao Bin He,et al.  Variable MWPCA for Adaptive Process Monitoring , 2008 .

[16]  Chai Tianyou.  Intelligent monitoring and control of mill load for grinding processes , 2008 .

[17]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[18]  Bogdan Gabrys,et al.  Data-driven Soft Sensors in the process industry , 2009, Comput. Chem. Eng..

[19]  Jialin Liu,et al.  On-line soft sensor for polyethylene process with multiple production grades , 2007 .

[20]  Barry M. Wise,et al.  Development and Benchmarking of Multivariate Statistical Process Control Tools for a Semiconductor Etch Process: Improving Robustness through Model Updating , 1997 .