Multiple model based soft sensor development with irregular/missing process output measurement

In this paper, nonlinear soft sensor development with irregular/missing output data is considered and a multiple model based modeling scheme is proposed for nonlinear processes. The efficiency of the proposed algorithm is demonstrated through several numerical simulation examples as well as the experimental data collected from a pilot-scale setup. It is shown through the comparison with the traditional missing data treatment methods in terms of the parameter estimation accuracy that, the developed soft sensors enjoy improved performance by employing the expectation-maximization (EM) algorithm in handling the missing process data and model varying problem.

[1]  Xiao Fan Wang,et al.  Soft sensing modeling based on support vector machine and Bayesian model selection , 2004, Comput. Chem. Eng..

[2]  Biao Huang,et al.  Dealing with Irregular Data in Soft Sensors: Bayesian Method and Comparative Study , 2008 .

[3]  Michael Athans,et al.  Guaranteed properties of gain scheduled control for linear parameter-varying plants , 1991, Autom..

[4]  Chonghun Han,et al.  A nonlinear soft sensor based on multivariate smoothing procedure for quality estimation in distillation columns , 2000 .

[5]  Bassam Bamieh,et al.  Identification of linear parameter varying models , 2002 .

[6]  Jialin Liu.  On-line soft sensor for polyethylene process with multiple production grades , 2005 .

[7]  Sirish L. Shah,et al.  Treatment of missing values in process data analysis , 2008 .

[8]  Manabu Kano,et al.  Inferential control system of distillation compositions using dynamic partial least squares regression , 2000 .

[9]  Ravindra D. Gudi,et al.  Identification of complex nonlinear processes based on fuzzy decomposition of the steady state space , 2003 .

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

[11]  R. Srinivasan,et al.  Data-driven Soft Sensor Approach For Quality Prediction in a Refinery Process , 2006, 2006 4th IEEE International Conference on Industrial Informatics.

[12]  J. Schafer,et al.  Missing data: our view of the state of the art. , 2002, Psychological methods.

[13]  Biao Huang,et al.  Robust identification of piecewise/switching autoregressive exogenous process , 2009 .

[14]  Bogdan Gabrys,et al.  Adaptive Local Learning Soft Sensor for Inferential Control Support , 2008, 2008 International Conference on Computational Intelligence for Modelling Control & Automation.

[15]  R. B. Gopaluni.  A particle filter approach to identification of nonlinear processes under missing observations , 2008 .

[16]  Raghuraj K. Rao,et al.  Genetic Programming Based Variable Interaction Models for Classification of Process and Biological Systems , 2009 .

[17]  D. Stoffer,et al.  Fitting Stochastic Volatility Models in the Presence of Irregular Sampling via Particle Methods and the EM Algorithm , 2008 .

[18]  Manabu Kano,et al.  Inferential control of distillation compositions: selection of model and control configuration , 2001 .

[19]  Yucai Zhu,et al.  Nonlinear MPC using an identified LPV model , 2009 .

[20]  J. Schafer,et al.  A comparison of inclusive and restrictive strategies in modern missing data procedures. , 2001, Psychological methods.