Skip to content
Licensed Unlicensed Requires Authentication Published by De Gruyter June 3, 2020

Comparative Analysis of Machine Learning Models for Predictive Control of the Cyclopentadine Production Process

Samuel V. Saraiva, Flávio V. Silva and Frede O. Carvalho


Different control strategies have been investigated to improve nonlinear system operations. One such strategy is the use of nonlinear predictive controllers (NMPCs) based on machine learning models. These models, such as artificial neural networks (NN), support vector machines (SVMs), and neuro-fuzzy networks (NF), present satisfactory adaptability to the complexity of the processes. In this aspect, a comparative study of the models in the predictive control of a complex system, such as MIMO (multiple-input-multiple-output) process of the production process of cyclopentadiene, is of interest and is the aim of this work. In this aspect, we find, through simulations, that the NMPCs presented adequate performance, especially those based on an SVM, concerning the servo and regulatory problem scenarios, keeping the process at the optimum operating point, especially for unattainable setpoint. The instability in the use of the classical proportional-integral-derivative linear control is also shown.

Funding statement: Coordenação de Aperfeiçoamento de Nível Superior – Brasil (CAPES), Grant Number: Finance Code 001.

A Appendices

The tabulated values for the parameters of cyclopentadiene process are shown in Table 5.

Table 5:

Parameters and variables of the studied process.

Kinetic Reaction Parameters
Arrhenius constant for reaction 1k01 = 1.287.1012 h−1
Arrhenius constant for reaction 2k02 = 1.287.1012 h−1
Arrhenius constant for reaction 3k03 = 9.043.109 l.h−1. mol−1
Activation Energy reaction 1E1/R = 9758.3 K
Activation Energy reaction 2E2/R = 9758.3 K
Activation Energy reaction 3E3/R = 8560 K
Reaction Enthalpy 1h1 = 4.2 kJ.mol−1
Reaction Enthalpy 1h2 = −11 kJ.mol−1
Reaction Enthalpy 1h3 = −41.85 kJ.mol−1
Physico-chemical parameters
Density0.9342 kg.l−1
Reagent heat capacityCpr = 3.01−1. K−1
Heat capacity of the cooling liquidCpc = 2.0−1. K−1
Thermal exchange coefficientU = 4032 kJ.h−1. m−2 K−1
Dimensions of the reactor
Thermal exchange areaAr = 0.215 m2
Reactor volumeVr = 10 l
Cooling liquid massmc = 5 kg

  1. Reference: [23]


[1] García CE, Prett DM, Morari M. Model predictive control: theory and practice—A survey. Automatica. 1989;25:335–48. DOI: 10.1016/0005-1098(89)90002-2.Search in Google Scholar

[2] Kouvaritakis B, Cannon M. Model Predictive Control, 1st ed. New York: Springer International Publishing. (Chapter 1). 2016.10.1007/978-3-319-24853-0Search in Google Scholar

[3] Forbes MG, Patwardhan RS, Hamadah H, Gopaluni RB. Model predictive control in industry: challenges and opportunities. IFAC-PapersOnLine, 9th IFAC Symp Adv Control Chem Processes ADCHEM 2015. 2015;48:531–8. DOI: 10.1016/j.ifacol.2015.09.022.Search in Google Scholar

[4] Kozák Š. State-of-the-art in control engineering. J Electr Syst Inf Technol. 2014;1:1–9. DOI: 10.1016/j.jesit.2014.03.002.Search in Google Scholar

[5] Grune L, Pannek J. Nonlinear model predictive control: theory and algorithms, 1st ed., (Chapter 3). London: Springer-Verlag, 2016.Search in Google Scholar

[6] Bouafia A, Gaubert J, Krim F. Design and implementation of predictive current control of three-phase PWM rectifier using space-vector modulation (SVM). Energy Convers Manage. 2010;51:2473–81. DOI: 10.1016/j.enconman.2010.05.010.Search in Google Scholar

[7] Sharma N, Singh K. Neural network and support vector machine predictive control oftert-amyl methyl ether reactive distillation column. Syst Sci Control Eng. 2014;2:512–26. DOI: 10.1080/21642583.2014.924082.Search in Google Scholar

[8] Wu Z, Tran A, Rincon D, Christofides PD. Machine learning-based predictive control of nonlinear processes. Part I: heory. AIChE 2019. 2019a;65:e16729.10.1002/aic.16729Search in Google Scholar

[9] Afram A, Janabi-Sharifi F, Fung SA, Raahemifar K. Artificial neural network (ANN) based model predictive control (MPC) and optimization of HVAC systems: A state of the art review and case study of a residential HVAC system. Energy Build. 2017;141:93–113. DOI: 10.1016/j.enbuild.2017.02.012.Search in Google Scholar

[10] Nagy ZK. Model based control of a yeast fermentation bioreactor using optimally designed artificial neural networks. Chem Eng J. 2007;127:95–109. DOI: 10.1016/j.cej.2006.10.015.Search in Google Scholar

[11] Yan Z, Wang J. Robust model predictive control of nonlinear systems with unmodeled dynamics and bounded uncertainties based on neural networks. IEEE Trans Neural Networks Learn Syst. 2014;25:457–69. DOI: 10.1109/TNNLS.2013.2275948.Search in Google Scholar

[12] Wu Z, Tran A, Rincon D, Christofides PD. Machine-learning-based predictive control of nonlinear processes. Part II: computational implementation. AIChE J. 2019b;65:e16734.10.1002/aic.16734Search in Google Scholar

[13] Zhongda T, Shujiang L, Yanhong W, Xiangdong W. SVM predictive control for calcination zone temperature in lime rotary kiln with improved PSO algorithm. Trans Inst Meas Control. 2018a;40:3134–46.10.1177/0142331217716983Search in Google Scholar

[14] Zhongda T, Shujiang L, Yanhong W, Xiangdong W. Mixed-kernel least square support vector machine predictive control based on improved free search algorithm for nonlinear systems. Trans Inst Meas Control. 2018b;40:4382–96.10.1177/0142331217748193Search in Google Scholar

[15] Zhongda T, Li S, Wang Y. T-S fuzzy neural network predictive control for burning zone temperature in rotary kiln with improved hierarchical genetic algorithm. Int J Model Identif Control. 2016;25:323–34.Search in Google Scholar

[16] Zhongda T, Li S, Wang Y. Generalized predictive PID control for main steam temperature based on improved PSO algorithm. J Adv Comput Intell Intell Inform. 2017;21:507–17.10.20965/jaciii.2017.p0507Search in Google Scholar

[17] Fatemeh Zahedi F, Zahedi Z. A review of neuro-fuzzy systems based on intelligent control. J Electr Electron Eng. 2015;3:58–61. DOI: 10.11648/j.jeee.s.2015030201.23.Search in Google Scholar

[18] Hopgood AA. Intelligent systems for engineers and scientists, 3th ed. New York: CRC Press. (Chapter 15). 2016.10.1201/b11287Search in Google Scholar

[19] Huang G, Huang G-B, Song S, You K. Trends in extreme learning machines: a review. Neural Networks. 2015;61:32–48. DOI: 10.1016/j.neunet.2014.10.001.Search in Google Scholar

[20] Ramirez H, Le Gorrec Y. Control of non-isothermal chemical reaction networks using irreversible port-Hamiltonian systems. IFAC-Papers OnLine. 2017;50:576–81. DOI: 10.1016/j.ifacol.2017.08.079.Search in Google Scholar

[21] Cassol GO, Campos GV, Thomaz DM, Capron BD, Secchi AR. Reinforcement learning applied to process control: a Van der Vusse Reactor case study. 13th Int Symp Process Syst Eng (PSE 2018). 2018;44:553–8. DOI: 10.1016/B978-0-444-64241-7.50087-2.Search in Google Scholar

[22] Ruchika, RN. Analysis of Van de Vusse Reactor using model predictive control. Int J Emerging Technol Adv Eng. 2013;3:236–40.Search in Google Scholar

[23] Vojtěšek I. Chemical reactors: modern control methods. Phd Thesis Czech Republic: Faculty of Applied Informatics,Tomas Bata University in Zlín,Zlín, 2008.Search in Google Scholar

[24] Gardezi MS, Hasan A. Machine learning based adaptive prediction horizon in finite control set model predictive control. IEEE Access. 2018;6:32392–400. DOI: 10.1109/ACCESS.2018.2839519.Search in Google Scholar

[25] Wu Z, Christofides PD. Optimizing process economics and operational safety via economic MPC using barrier functions and recurrent neural network models. Chem Eng Res Des. 2019c;2019:ISSN 0263-8762.10.1016/j.cherd.2019.10.010Search in Google Scholar

[26] Haykin S. Neural networks and learning machines, 3th ed., (Chapter 4,6). US: Pearson Education, 2008.Search in Google Scholar

[27] Jang J-S. ANFIS: adaptive-network-based fuzzy inference system. IEEE Trans Syst Man Cybern. 1993;23:665–85. DOI: 10.1109/21.256541.Search in Google Scholar

[28] Sugeno M, Kang G. Structure identification of fuzzy model. Fuzzy Sets Syst. 1988;28:15–33. DOI: 10.1016/0165-0114(88)90113-3.Search in Google Scholar

[29] Karaboga D, Kaya E. Adaptive network based fuzzy inference system (ANFIS) training approaches: a comprehensive survey. Artif Intell Rev. 2018;1–31. doi:10.1007/s10462-017-9610–2.Search in Google Scholar

[30] Vapnik VN. The nature of statistical learning theory. 2nd ed. New York: Springer Verlag, 1999.10.1007/978-1-4757-3264-1Search in Google Scholar

[31] Wang L. Support vector machines: theory and applications. 1st ed. Berlin Heidelberg: Springer-Verlag, 2005.10.1007/b95439Search in Google Scholar

[32] Zhongda Tian, Shujiang Li, Yanhong Wang, Xiangdong Wang. Mixed-kernel least square support vector machine predictive control based on improved free search algorithm for nonlinear systems. Transactions of the Institute of Measurement and Control. 2018 2;40(16):4382–4396.10.1177/0142331217748193.Search in Google Scholar

Received: 2019-09-16
Revised: 2019-11-19
Accepted: 2019-11-22
Published Online: 2020-06-03

© 2020 Walter de Gruyter GmbH, Berlin/Boston