Probabilistic learning of partial least squares regression model: Theory and industrial applications

Abstract This paper formulates a probabilistic form of the widely used Partial Least Squares (PLS) model for regression modeling and application in industrial processes. Different from the existing probabilistic Principal Component Analysis/Principal Component Regression models, two types of latent variables are introduced into the probabilistic PLS model structure. For training and parameter learning of the probabilistic PLS model, the Bayes rule is applied and an efficient Expectation-Maximization algorithm is designed. Furthermore, in order to describe more complicated processes, the single probabilistic PLS model is extended to the mixture form under a similar probabilistic modeling framework. Two industrial case studies are provided as examples of the application of soft sensors constructed based on the new developed models.

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

[2]  Morimasa Ogawa,et al.  The state of the art in chemical process control in Japan: Good practice and questionnaire survey , 2010 .

[3]  Kaixiang Peng,et al.  A comparison and evaluation of key performance indicator-based multivariate statistics process monitoring approaches ☆ , 2015 .

[4]  Biao Huang,et al.  A Bayesian approach to design of adaptive multi-model inferential sensors with application in oil sand industry , 2012 .

[5]  Baoju Zhang,et al.  A Unified Probabilistic PLSR Model for Quantitative Analysis of Surface-Enhanced Raman Spectrum (SERS) , 2014, ICC 2014.

[6]  Haiqing Wang,et al.  Soft Chemical Analyzer Development Using Adaptive Least-Squares Support Vector Regression with Selective Pruning and Variable Moving Window Size , 2009 .

[7]  Luigi Fortuna,et al.  Soft sensors for product quality monitoring in debutanizer distillation columns , 2005 .

[8]  Michael E. Tipping,et al.  Probabilistic Principal Component Analysis , 1999 .

[9]  Biao Huang,et al.  Design of inferential sensors in the process industry: A review of Bayesian methods , 2013 .

[10]  Jean X. Gao,et al.  Probabilistic Partial Least Square Regression: A Robust Model for Quantitative Analysis of Raman Spectroscopy Data , 2011, 2011 IEEE International Conference on Bioinformatics and Biomedicine.

[11]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[12]  Jin Wang,et al.  Comparison of the performance of a reduced-order dynamic PLS soft sensor with different updating schemes for digester control , 2012 .

[13]  Ali Cinar,et al.  Monitoring, fault diagnosis, fault-tolerant control and optimization: Data driven methods , 2012, Comput. Chem. Eng..

[14]  In-Beum Lee,et al.  Process monitoring based on probabilistic PCA , 2003 .

[15]  Si-Zhao Joe Qin,et al.  Survey on data-driven industrial process monitoring and diagnosis , 2012, Annu. Rev. Control..

[16]  Jie Yu,et al.  Online quality prediction of nonlinear and non-Gaussian chemical processes with shifting dynamics using finite mixture model based Gaussian process regression approach , 2012 .

[17]  Yi Liu,et al.  Integrated soft sensor using just-in-time support vector regression and probabilistic analysis for quality prediction of multi-grade processes , 2013 .

[18]  Furong Gao,et al.  Mixture probabilistic PCR model for soft sensing of multimode processes , 2011 .

[19]  Furong Gao,et al.  Review of Recent Research on Data-Based Process Monitoring , 2013 .

[20]  Soon Keat Tan,et al.  Localized, Adaptive Recursive Partial Least Squares Regression for Dynamic System Modeling , 2012 .

[21]  Mats G. Gustafsson,et al.  A Probabilistic Derivation of the Partial Least-Squares Algorithm , 2001, J. Chem. Inf. Comput. Sci..

[22]  Tao Chen,et al.  Robust probabilistic PCA with missing data and contribution analysis for outlier detection , 2009, Comput. Stat. Data Anal..