Ensemble local kernel learning for online prediction of distributed product outputs in chemical processes

Abstract The crystal size distribution in crystallization processes, the molecular weight distribution in polymerization processes, the particle size distribution in powder industries, and the pulp fiber length distribution in paper industries are all distributed product outputs. Reliable online quality prediction of these chemical processes with distributed outputs is important but challenging. In this work, the kernel learning (KL) framework is introduced to model and online predict the distributed product outputs. First, the KL method is proposed to construct a global distributed shape. Then, without resorting to a KL-based global distributed model, a just-in-time KL (JKL) model is presented for better description of local distributed shapes with more accurate and quick prediction performance. Moreover, an ensemble JKL (EJKL) modeling approach is developed to obtain more reliable prediction performance of the distributed outputs. The proposed modeling methods are applied to online prediction of the molecular weight distribution in polymerization processes and of the crystal size distribution in crystallization processes. The prediction results show the proposed method is superior to the traditional counterparts.

[1]  Lei Xie,et al.  Inferential Model for Industrial Polypropylene Melt Index Prediction with Embedded Priori Knowledge and Delay Estimation , 2012 .

[2]  R. Braatz,et al.  Particle Size and Shape Control in Crystallization Processes , 2002 .

[3]  John F. MacGregor,et al.  Optimization of molecular-weight distribution using batch-to-batch adjustments , 1998 .

[4]  Lei Guo,et al.  Generalized discrete-time PI control of output PDFs using square root B-spline expansion , 2005, Autom..

[5]  Richard D Braatz,et al.  Advances and new directions in crystallization control. , 2012, Annual review of chemical and biomolecular engineering.

[6]  Liu Yi,et al.  Online Identification of Time-varying Processes Using Just-in-time Recursive Kernel Learning Approach , 2014 .

[7]  J. Romagnoli,et al.  On-line multi-variable predictive control of molar mass and particle size distributions in free-radical emulsion copolymerization , 2005 .

[8]  Jeffrey D. Ward,et al.  The best objective function for seeded batch crystallization , 2013 .

[9]  S. Shioya,et al.  Molecular weight distribution control in a batch polymerization reactor , 1988 .

[10]  Yi Hu,et al.  Enhanced batch process monitoring using just-in-time-learning based kernel partial least squares , 2013 .

[11]  Lei Xie,et al.  Novel Just-In-Time Learning-Based Soft Sensor Utilizing Non-Gaussian Information , 2014, IEEE Transactions on Control Systems Technology.

[12]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machines , 2002 .

[13]  Zainal Ahmad,et al.  Modelling and control of different types of polymerization processes using neural networks technique: A review , 2010 .

[14]  Zoltan K. Nagy,et al.  Model based robust control approach for batch crystallization product design , 2009, Comput. Chem. Eng..

[15]  Gary A. Smook,et al.  Handbook for Pulp and Paper Technologists , 1982 .

[16]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[17]  J. Leiza,et al.  Maximizing production and polymer quality (MWD and composition) in emulsion polymerization reactors with limited capacity of heat removal , 2003 .

[18]  Christos Georgakis,et al.  Data-driven, using design of dynamic experiments, versus model-driven optimization of batch crystallization processes , 2013 .

[19]  J. Romagnoli,et al.  Online control of molar mass and particle‐size distributions in emulsion polymerization , 2006 .

[20]  M. Birattari,et al.  Lazy learning for local modelling and control design , 1999 .

[21]  Alexander J. Smola,et al.  Learning with Kernels: support vector machines, regularization, optimization, and beyond , 2001, Adaptive computation and machine learning series.

[22]  Andrew W. Moore,et al.  Locally Weighted Learning , 1997, Artificial Intelligence Review.

[23]  T. Crowley,et al.  Experimental studies on optimal molecular weight distribution control in a batch-free radical polymerization process , 1998 .

[24]  Young-Don Ko,et al.  A neural network-based soft sensor for particle size distribution using image analysis , 2011 .

[25]  Costas Kiparissides,et al.  Challenges in particulate polymerization reactor modeling and optimization: A population balance perspective , 2006 .

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

[27]  Yi Liu,et al.  Industrial melt index prediction with the ensemble anti-outlier just-in-time Gaussian process regression modeling method , 2015 .

[28]  D. L. Ma,et al.  Optimal seeding in batch crystallization , 1999 .

[29]  Francis J. Doyle,et al.  Control of particle size distribution described by a population balance model of semibatch emulsion polymerization , 2000 .

[30]  Manabu Kano,et al.  Soft‐sensor development using correlation‐based just‐in‐time modeling , 2009 .

[31]  Wei Tang,et al.  Ensembling neural networks: Many could be better than all , 2002, Artif. Intell..

[32]  Michel Perrier,et al.  Shaping probability density functions using a switching linear controller , 2011 .

[33]  Francis J. Doyle,et al.  Open-loop control of particle size distribution in semi-batch emulsion copolymerization using a genetic algorithm , 2002 .

[34]  Jyh-Shyong Chang,et al.  Molecular weight control of a batch polymerization reactor : Experimental study , 1999 .

[35]  Hyun-Chul Kim,et al.  Constructing support vector machine ensemble , 2003, Pattern Recognit..

[36]  Zengliang Gao,et al.  Just-in-time kernel learning with adaptive parameter selection for soft sensor modeling of batch processes , 2012 .

[37]  Ping Li,et al.  Selective recursive kernel learning for online identification of nonlinear systems with NARX form , 2010 .

[38]  Jie Zhang,et al.  Inferential Estimation of Polymer Melt Index Using Sequentially Trained Bootstrap Aggregated Neural Networks , 2006 .

[39]  Hong Wang.  Minimum entropy control of non-Gaussian dynamic stochastic systems , 2002, IEEE Trans. Autom. Control..

[40]  Richard D. Braatz,et al.  Advanced control of crystallization processes , 2002, Annu. Rev. Control..

[41]  Hong Wang,et al.  Suboptimal mean controllers for bounded and dynamic stochastic distributions , 2002 .

[42]  Gavin C. Cawley,et al.  Fast exact leave-one-out cross-validation of sparse least-squares support vector machines , 2004, Neural Networks.

[43]  John R. Richards,et al.  Measurement and control of polymerization reactors , 2006, Comput. Chem. Eng..

[44]  Jianhua Zhang,et al.  Shaping of molecular weight distribution by iterative learning probability density function control strategies , 2008 .

[45]  M. Chiu,et al.  A new data-based methodology for nonlinear process modeling , 2004 .

[46]  Lei Guo,et al.  PID controller design for output PDFs of stochastic systems using linear matrix inequalities , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[47]  Panagiotis D. Christofides,et al.  Predictive control of particle size distribution in particulate processes , 2006 .

[48]  Jianhua Zhang,et al.  Bounded stochastic distributions control for pseudo-ARMAX stochastic systems , 2001, IEEE Trans. Autom. Control..

[49]  Hong Wang,et al.  Optimal probability density function control for NARMAX stochastic systems , 2008, Autom..

[50]  J. Macgregor,et al.  Control of Particle Size Distributions in Emulsion Semibatch Polymerization Using Mid-Course Correction Policies , 2002 .

[51]  W. Harmon Ray,et al.  The dynamic behavior of continuous polymerization reactors—I: Isothermal solution polymerization in a CSTR , 1981 .

[52]  Lei Guo,et al.  Minimum entropy filtering for multivariate stochastic systems with non-Gaussian noises , 2005, Proceedings of the 2005, American Control Conference, 2005..

[53]  Yi Liu,et al.  Development of soft-sensors for online quality prediction of sequential-reactor-multi-grade industrial processes , 2013 .