Integration of Multivariable Model Predictive Control, Statistical Process Con-trol, and Artificial Neural Networks for the Detection and Management of Structural Changes in Industrial MIMO Processes
DOI:
https://doi.org/10.17268/scien.inge.2026.02.04Keywords:
multivariable model predictive control, statistical process control, artificial neural networks, MIMO processes, process supervisionAbstract
This research addresses the detection and monitoring of structural changes in multivariable MIMO industrial processes through an integrated architecture that combines multivariable model predictive control, statistical process control, and artificial neural networks. The objective is to improve operational discrimination among external disturbances, expected operating conditions, and internal structural changes in the process, while preserving adequate closed-loop regulation performance. The methodology is based on controlled and reproducible simulations of a MIMO process under systematically designed scenarios, including step disturbances, activation of operational constraints, and internal parameter drifts. Multivariable model predictive control acts as the regulation mechanism, whereas statistical process control supervises statistical stability through multivariate indicators applied to model residuals. Artificial neural networks are employed as nonlinear reference models for structural consistency assessment, without directly intervening in the control action. The results show that the MPC+ANN+MSPC architecture reduced the post-event alarm occupancy from 0.9834 to 0.1492 in Scenario S5, while maintaining RMSE, MAE, and control effort values comparable to those of the linear MSPC scheme. It is concluded that the proposed integration is a robust and viable tool for advanced supervision of MIMO industrial processes.
References
Castelletti, A., Ficchì, A., Cominola, A., Segovia, P., Giuliani, M., Wu, W., Lucia, S., Ocampo-Martinez, C., De Schutter, B., & Maestre, J. M. (2023). Model predictive control of water resources systems: A review and research agenda. Annual Reviews in Control, 55, 442–465. https://doi.org/10.1016/j.arcontrol.2023.03.013
Cavalcanti, F., Kozonoe, C. E., Pacheco, K., & Alves, R. M. B. (2021). Application of artificial neural net-works to chemical and process engineering. In Deep learning applications. IntechOpen. https://doi.org/10.5772/intechopen.96641
Fiedler, F., Karg, B., Lüken, L., Brandner, D., Heinlein, M., Brabender, F., & Lucia, S. (2023). do-mpc: To-wards FAIR nonlinear and robust model predictive control. Control Engineering Practice, 140, 105676. https://doi.org/10.1016/j.conengprac.2023.105676
Hernandez, C., Matta, N., Almadhoun, W., Chelligue, T., Palomino, C., Alhammo, A., Ayoubi, R., Arnaout, A., Penubarthi, S., Pandey, M., Abdullayev, A., Sabbagh, A. A., Asarpota, J., Mijares, G., & Damarla, P. (2024). Revolutionizing oil field operations: Implementing multivariable predictive control to drive au-tonomous well control operations. ADIPEC. https://doi.org/10.2118/222967-MS
Herrera Vega, J. C., de Jesús Rahmer, B., & Herrera Vidal, G. (2022). Análisis de componentes principales (ACP) aplicados al control estadístico de procesos multivariados. Investigación e Innovación en Inge-nierías, 10(1), 17–29. https://doi.org/10.17081/invinno.10.1.4972
Huang, K., Wei, K., Li, F., Yang, C., & Gui, W. (2023). LSTM-MPC: A deep learning based predictive control method for multimode process control. IEEE Transactions on Industrial Electronics, 70(11), 11544–11554. https://doi.org/10.1109/TIE.2022.3229323
Jung, M., Mendes, P. R. d. C., Önnheim, M., & Gustavsson, E. (2023). Model predictive control when utilizing LSTM as dynamic models. Engineering Applications of Artificial Intelligence, 123, 106226. https://doi.org/10.1016/j.engappai.2023.106226
Li, J., Ding, D., & Tsung, F. (2021). Directional PCA for fast detection and accurate diagnosis: A unified framework. IEEE Transactions on Cybernetics. https://doi.org/10.1109/TCYB.2021.3070590
Melo, A., Câmara, M., & Pinto, J. C. (2024). Data-driven process monitoring and fault diagnosis: A compre-hensive survey. Processes, 12(2), 251. https://doi.org/10.3390/pr12020251
Ren, Y. M., Alhajeri, M. S., Luo, J., Chen, S., Abdullah, F., Wu, Z., & Christofides, P. D. (2022). A tutorial re-view of neural network modeling approaches for model predictive control. Computers & Chemical Engi-neering, 165, 107956. https://doi.org/10.1016/j.compchemeng.2022.107956
Rivero-Contreras, R. E., Zamarreño, J. M., & Tadeo, F. (2025). Avances en el control predictivo para con-troladores lógicos programables. Revista Iberoamericana de Automática e Informática Industrial, 23(1), 13–24. https://doi.org/10.4995/riai.2025.22466
Schwenzer, M., Ay, M., Bergs, T., & Abel, D. (2021). Review on model predictive control: An engineering perspective. The International Journal of Advanced Manufacturing Technology, 117(5–6), 1327–1349. https://doi.org/10.1007/s00170-021-07682-3
Ueda, R. M., & Souza, A. M. (2022). An effective approach to detect the source(s) of out-of-control signals in productive processes by vector error correction (VEC) residual and Hotelling’s T² decomposition tech-niques. Expert Systems with Applications, 187, 115979. https://doi.org/10.1016/j.eswa.2021.115979
Wang, K., Chen, J., Song, Z., Wang, Y., & Yang, C. (2022). Deep neural network-embedded stochastic non-linear state-space models and their applications to process monitoring. IEEE Transactions on Neural Networks and Learning Systems, 33(12), 7682–7694. https://doi.org/10.1109/TNNLS.2021.3086323
Yu, Z., & Long, J. (2024). Review on advanced model predictive control technologies for high-power con-verters and industrial drives. Electronics, 13(24), 4969. https://doi.org/10.3390/electronics13244969
Zheng, J., Ma, L., Wu, Y., Ye, L., & Shen, F. (2022). Nonlinear dynamic soft sensor development with a su-pervised hybrid CNN–LSTM network for industrial processes. ACS Omega, 7(19), 16653–16664. https://doi.org/10.1021/acsomega.2c01108
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Los autores/as que publiquen en esta revista aceptan las siguientes condiciones:
- Los autores/as conservan los derechos de autor y ceden a la revista el derecho de la primera publicación, con el trabajo registrado con la licencia de atribución de Creative Commons, que permite a terceros utilizar lo publicado siempre que mencionen la autoría del trabajo y a la primera publicación en esta revista.
- Los autores/as pueden realizar otros acuerdos contractuales independientes y adicionales para la distribución no exclusiva de la versión del artículo publicado en esta revista (p. ej., incluirlo en un repositorio institucional o publicarlo en un libro) siempre que indiquen claramente que el trabajo se publicó por primera vez en esta revista.
- Se permite y recomienda a los autores/as a publicar su trabajo en Internet (por ejemplo en páginas institucionales o personales) antes y durante el proceso de revisión y publicación, ya que puede conducir a intercambios productivos y a una mayor y más rápida difusión del trabajo publicado
