Performance Comparison of Different Regression Methods for a Polymerization Process with Adaptive Sampling

Developing complete mechanistic models for polymerization reactors is not easy, because complex reactions occur simultaneously; there is a large number of kinetic parameters involved and sometimes the chemical and physical phenomena for mixtures involving polymers are poorly understood. To overcome these difficulties, empirical models based on sampled data can be used instead, namely regression methods typical of machine learning field. They have the ability to learn the trends of a process without any knowledge about its particular physical and chemical laws. Therefore, they are useful for modeling complex processes, such as the free radical polymerization of methyl methacrylate achieved in a batch bulk process. The goal is to generate accurate predictions of monomer conversion, numerical average molecular weight and gravimetrical average molecular weight. This process is associated with non-linear gel and glass effects. For this purpose, an adaptive sampling technique is presented, which can select more samples around the regions where the values have a higher variation. Several machine learning methods are used for the modeling and their performance is compared: support vector machines, k-nearest neighbor, k-nearest neighbor and random forest, as well as an original algorithm, large margin nearest neighbor regression. The suggested method provides very good results compared to the other well-known regression algorithms.

Authors:

• Florin Leon
• Silvia Curteanu

Keywords:

• Adaptive sampling
• batch bulk methyl methacrylate polymerization
• large margin nearest neighbor regression
• machine learning.

References:

[1] S. Curteanu, “Direct and Inverse Neural Network Modeling in Free Radical Polymerization”, Central European Journal of Chemistry,
vol. 2, no. 1, 2004, pp. 113–140.
[2] S. Curteanu, F. Leon and D. Gâlea, “Neural Network Models for Free Radical Polymerization of Methyl Methacrylate”, Eurasian Chemico-Technological Journal, vol. 5, no. 3, 2003, pp. 225–231.
[3] J. C. B. Gonzaga, L. A. C. Meleiro, C. Kiang and R. Maciel Filho, “ANN-based soft-sensor for real-time process monitoring and control of an industrial polymerization process”, Computers & Chemical Engineering, vol. 33, no. 1, 2009, pp. 43–49.
[4] S. Contant, P. V. R. Mesa and L. M. F. Lona, “Modeling of Styrene Living Radical Polymerization Mediated by Tempo Using Neural Networks”, Proceedings of the 2nd Mercosur Congress on Chemical Engineering and 4th Mercosur Congress on Process System Engineering, ENPROMER, 2005, Rio de Janeiro, Brazil, 2005, pp. 1–10.
[5] F. A. N. Fernandes, “Selection of a mixture of initiators for batch polymerization using neural networks”, Journal of Applied Polymer Science, vol. 98, issue 5, 2005, pp. 2088–2093.
[6] S. Curteanu, F. Leon, R. Furtună, E. N. Drăgoi and N. Curteanu, “Comparison between Different Methods for Developing Neural Network Topology Applied to a Complex Polymerization Process”, Proceedings of the International Joint Conference on Neural Networks, IEEE World Congress on Computational Intelligence, Barcelona, Spain, 2010, pp. 1293–1300.
[7] M. S. Leite, B. F. Dos Santos, L. M. F. Lona, F. V. Da Silva and A. M. Frattini Fileti, “Application of Artificial Intelligence Techniques for Temperature Prediction in a Polymerization Process”, Chemical Engineering Transactions, vol. 24, 2011, pp. 385–390.
[8] A. Salman, A. P. Engelbrecht and M. G. H. Omran, “Empirical analysis of self-adaptive differential evolution”, European Journal of Operational Research, vol. 183, issue 2, 2007, pp. 785–804.
[9] M. Hall, E. Frank, G. Holmes, B. Pfahringer, P. Reutemann and I. H. Witten, “The WEKA Data Mining Software: An Update”, ACM SIGKDD Explorations, vol. 11, no. 1, 2009, pp. 10–18.
[10] F. Leon and S. Curteanu, “Evolutionary Algorithm for Large Margin Nearest Neighbour Regression”, Proceedings of the 7th International Conference on Computational Collective Intelligence Technologies and Applications, ICCCI 2015, Madrid, Spain, Part I, Lecture Notes in Artificial Intelligence, LNAI 9329, Springer International Publishing Switzerland, doi: 10.1007/978-3-319-24069-5_29, 2015, pp. 286–296.
[11] F. Leon and S. Curteanu, “Large Margin Nearest Neighbour Regression Using Different Optimization Techniques”, Journal of Intelligent and Fuzzy Systems, IOS Press, in press
[12] K. Q. Weinberger, J. Blitzer and L. K. Saul, “Distance metric learning for large margin nearest neighbor classification”, Advances in Neural Information Processing Systems, vol. 18, MIT Press, Cambridge, MA, USA, 2006, pp. 1473–1480.
[13] K. Q. Weinberger and L. K. Saul, “Fast solvers and efficient implementations for distance metric learning”, Proceedings of the 25th International Conference on Machine Learning, Helsinki, Finland, 2008, pp. 1160–1167.
[14] K. Q. Weinberger and L. K. Saul, “Distance Metric Learning for Large Margin Nearest Neighbor Classification”, Journal of Machine Learning Research, vol. 10, 2009, pp. 207–244.
[15] H. Jeffreys and B. S. Jeffreys, “Central Differences Formula”, in Methods of Mathematical Physics, 3rd edition, Cambridge University Press, 1988, pp. 284–286.