A Model-Based Control Strategy for Production of High Acrylonitrile Content Nitrile-Butadiene Rubber (NBR)
DOI:
https://doi.org/10.70567/mc.v41i16.87Keywords:
Soft sensors, batch processes, recurrent neuronal networks, NBR rubberAbstract
High A-content NBR (nitrile-butadiene rubber) is typically produced through emulsion copolymerization of acrylonitrile and butadiene. Production is carried out above the “azeotropic point,” where the process can become unstable, hindering uniform product quality. Limitations in online copolymer composition measurements restrict closed-loop control strategies for process stability. This study proposes a closed-loop control strategy to adjust copolymer composition during operation above the azeotropic point. Based on a first-principles model, a recurrent neural network inferential sensor estimates composition online, enabling closed-loop control via B dosing throughout the process. Results demonstrate acceptable control methodology performance, ensuring stable conditions and uniform composition, even with significant modeling errors.
References
Clementi, L. A., Suvire, R. B., Rossomando, F.G., and Vega, J.R., A Closed-Loop Control Strategy for Producing Nitrile Rubber of Uniform Chemical Composition in a Semibatch Reactor: A Simulation Study. Macro-molecular Reaction Engineering, 12:1700054, 2018. https://doi.org/10.1002/mren.201700054
Grimm, D. C. (1998). Method for the Production of Nitrile rubber. U.S. Patent: No. 5708132.
Ji, C., Ma, F., Wang, J., and Sun W., Profitability related industrial-scale batch processes monitoring via deep learning based soft sensor development. Computers & Chemical Engineering, 170:108125, 2023. https://doi.org/10.1016/j.compchemeng.2022.108125
Lei, Y., Karimi, H. R., and Chen, X., A novel self-supervised deep LSTM network for industrial temperature prediction in aluminum processes application. Neurocomputing, 502:177-185, 2022. https://doi.org/10.1016/j.neucom.2022.06.080
Lightsey, J. W. (1998). Continuous Polymerization Process for Producing NBR Rubber having a High Bound Content of Acrylontrile. U.S. Patent: No. 5770660.
Mowbray, M., Kay, H., Kay, S., Castro Caetano, P., Hicks, A., Mendoza, C., Lane, A., Martin, P., and Zhang, D., Probabilistic machine learning based soft-sensors for product quality prediction in batch processes. Chemometrics and Intelligent Laboratory Systems, 228:104616, 2022. https://doi.org/10.1016/j.chemolab.2022.104616
Qiu, K., Wang, J., Zhou, X., Wang, R., and Guo, Y., Soft sensor based on localized semisupervised relevance vector machine for penicillin fermentation process with asymmetric data. Measurement, 202:111823, 2022. https://doi.org/10.1016/j.measurement.2022.111823
Qiu, K., Wang, J., Wang, R., Guo, Y., and Zhao, L., Soft sensor development based on kernel dynamic time warping and a relevant vector machine for unequal-length batch processes. Expert Systems with Applications, 182:115223, 2021. https://doi.org/10.1016/j.eswa.2021.115223
Ren, J., Ni, D., A batch-wise LSTM-encoder decoder network for batch process monitoring. Chemical Engineering Research and Design, 164:102-112, 2020. https://doi.org/10.1016/j.cherd.2020.09.019
Shokry, A., Vicente, P., Escudero, G., Pérez-Moya, M., Graells, M., and Espuña, A., Datadriven soft-sensors for online monitoring of batch processes with different initial conditions. Computers & Chemical Engineering, 118:159-179, 2018. https://doi.org/10.1016/j.compchemeng.2018.07.014
Vega, J.R., Gugliotta, L., Bielsa, R., Brandolini, M., and Meira, G., Emulsion copolymerization of acrylonitrile and butadiene. Mathematical model of an industrial reactor. Ind. Eng. Chem. Res, 36:1238-1246, 1998. https://doi.org/10.1021/ie9605342
Downloads
Published
Issue
Section
License
Copyright (c) 2024 Argentine Association for Computational Mechanics
This work is licensed under a Creative Commons Attribution 4.0 International License.
This publication is open access diamond, with no cost to authors or readers.
Only those papers that have been accepted for publication and have been presented at the AMCA congress will be published.
