Parsimonious neural modeling for the control of lateritic mineral drying
Keywords:
industrial drying; artificial neural networks; system identification; automatic controlAbstract
Introduction: The thermal drying process of lateritic minerals is one of the most energy-intensive stages in the nickel metallurgy industry, with combustion efficiency being a determining factor in the overall system performance.
Objective: To develop artificial neural network models that describe the nonlinear dynamics of the control loops of the lateritic mineral drying process, using a multi-objective selection criterion that guarantees an optimal balance between predictive accuracy and structural parsimony.
Methods: The research was conducted using a passive experimental approach, employing real historical operating data corresponding to one month of continuous system operation. Multilayer perceptron architectures with different numbers of neurons in the hidden layer were evaluated, using the mean squared error and the Bayesian information criterion to select the optimal model.
Results: An architecture with four hidden neurons was identified as providing the best compromise between accuracy and parsimony. For the primary air loop, an average mean squared error of 0.79596 × 106 m³/h and a Bayesian information criterion of 3.4954 × 104 were obtained, while for the oil loop, the mean squared error was 1.1740 × 104 kg/h with a Bayesian information criterion of 2.4149 × 104.
Conclusions: The selected models demonstrate adequate predictive and generalization capabilities, as well as the relevance of the proposed approach to support the improvement of automatic control of the drying process, with potential impact on energy efficiency and operational stability.
Downloads
References
1. Angulo-Palma HJ, Salazar I, Saldana M, et al. New Model to Predict Nickel Extraction from Lateritic Ores During the Roasting–Reduction of the Caron Process. Minerals [Internet]. 2026 marzo [citado 21 may 2026];16(3):278. Disponible en: https://doi.org/10.3390/min16030278
2. Lv W, Makuza B, Wang F. A Review of Direct Reduction–Magnetic Separation Process for Ferronickel Production from Nickel Laterite. Journal of Sustainable Metallurgy [Internet]. 2025 noviembre [citado 21 may 2026];11:3-28. Disponible en: https://doi.org/10.1007/s40831-024-00950-y
3. Acquah G, Skinner W, Abaka-Wood G, et al. Extraction of Nickel and Cobalt from Complex Low-Grade Lateritic Ores: Challenges and Opportunities. Minerals [Internet]. 2026 marzo [citado 21 may 2026];16(3):287. Disponible en: https://doi.org/10.3390/min16030287
4. Zevgolis EN, Daskalakis KA. The Nickel Production Methods from Laterites and the Greek Ferronickel Production among Them. Materials Proceedings [Internet]. 2021 [citado 21 may 2026];5(1):104. Disponible en: https://doi.org/10.3390/materproc2021005104
5. Natural Resources Canada. Nickel facts [Internet]. Ottawa (ON): Government of Canada; 2025 [citado 28 mar 2026]. Disponible en: https://natural-resources.canada.ca/minerals-mining/mining-data-statistics-analysis/minerals-metals-facts/nickel-facts
6. Palma HJA, Toro N, Castillo J, et al. New Evidence of the Effect of Post-Combustion During Roasting/Reduction of Nickel-Bearing Lateritic Ores. Processes [Internet]. 2025 agosto [citado 21 may 2026];13(8):2602. Disponible en: https://doi.org/10.3390/pr13082602
7. Angulo-Palma HJ, Legrá Legrá Á, Urgellés AL, et al. Use of a Mixture of Coal and Oil as an Additive for Selective Reduction of Lateritic Ore by the Caron Process. Hemijska Industrija [Internet]. 2024 febrero [citado 21 may 2026]; 78(1):17-27. Disponible en: https://doi.org/10.2298/HEMIND230118017A
8. Rojas Vargas A, Magaña Haynes ME, del Toro Álvarez D, et al. Nickel Mining: Studying the Kinetics and Modeling of Extraction Process from Laterite Ore. South African Journal of Chemical Engineering [Internet]. 2025 julio [citado 21 may 2026];53:362-72. Disponible en: https://doi.org/10.1016/j.sajce.2025.06.002
9. Castaño F, Rubio FR, Ortega MG. Modelado de Secaderos Rotatorios en Isocorriente. Revista Iberoamericana de Automática e Informática Industrial [Internet]. 2009 octubre [citado 14 feb 2026];6(4):32-43. Disponible en: https://doi.org/10.1016/S1697-7912(09)70106-2
10. Fedorova ER, Morgunov VV, Pupysheva EA. Effect of variation of internal diameter along the length of a rotary kiln on material movement. Non-Ferrous Metals [Internet]. 2024 julio [citado 7 may 2026];56(1):28-34. Disponible en: https://doi.org/10.17580/nfm.2024.01.05
11. Fedorova E, Morgunov V, Lobko K, et al. Review: Axial Motion of Material in Rotary Kilns. Eng [Internet]. 2025 mayo [citado 7 may 2026];6(6):106. Disponible en: https://doi.org/10.3390/eng6060106
12. Kukharova T, Medvedeva E, Novozhilov I. Modeling of a Temperature Field Control System for a Carburizing Shaft Furnace. Processes [Internet]. 2026 abril [citado 7 may 2026];14(9):1406. Disponible en: https://doi.org/10.3390/pr14091406
13. Kukharova TV, Ilyukhina YA, Shestopalov MY. Development of a Methodology for Controlling the Process of Heating Metal Blanks in a Methodical Furnace. En: Proceedings of the 2022 Conference of Russian Young Researchers in Electrical and Electronic Engineering ElConRus [Internet]. 2022 abril [citado 7 may 2026];718-21. Disponible en: https://doi.org/10.1109/ElConRus54750.2022.9755574
14. Khan MIH, Batuwatta-Gamage CP, Karim MA, et al. Fundamental understanding of heat and mass transfer processes for physics-informed machine learning-based drying modelling. Energies [Internet]. 2022 diciembre [citado 21 mayo 2026];15(24):9347. Disponible en: https://doi.org/10.3390/en15249347
15. Bishnu SK, Alnouri SY, Al Mohannadi DM. Computational applications using data-driven modeling in process systems: A review. Digital Chemical Engineering [Internet]. 2023 septiembre [citado 21 may 2026];8:100111. Disponible en: https://doi.org/10.1016/j.dche.2023.100111
16. Rodríguez-Durán VG, Montero-Góngora D, Columbié-Navarro ÁO, et al. Low-order parametric modeling and control of air–fuel loops in industrial rotary dryers: A passive system identification approach. Drying Technology [Internet]. 2026 mar [citado 26 mar 2026];44(6):808-23. Disponible en: https://doi.org/10.1080/07373937.2026.2638948
17. Ren YM, Alhajeri MS, Luo J, et al. A tutorial review of neural network modeling approaches for model predictive control. Computers & Chemical Engineering [Internet]. 2022 septiembre [citado 22 mar 2026];165:107956. Disponible en: https://doi.org/10.1016/j.compchemeng.2022.107956
18. Rashid M, Mhaskar P. Are Neural Networks the Right Tool for Process Modeling and Control of Batch and Batch-like Processes? Processes [Internet]. 2023 enero [citado 11 ene 2026];11(3):686. Disponible en: https://doi.org/10.3390/pr11030686
19. Wang Z, Wang H, Du Z. Deep Neural Network Model Based on Process Mechanism Applied to Predictive Control of Distillation Processes. Processes [Internet]. 2025 febrero [citado 11 ene 2026];13(3):811. Disponible en: https://doi.org/10.3390/pr13030811
20. Carter A, Imtiaz S, Naterer GF. Review of interpretable machine learning for process industries. Process Safety and Environmental Protection [Internet]. 2023 febrero [citado 11 mar 2026];170:647-659. Disponible en: https://doi.org/10.1016/j.psep.2022.12.018
21. Dambros JWV, Trierweiler JO, Farenzena M, et al. Oscillation Detection in Process Industries by a Machine Learning-Based Approach. Industrial & Engineering Chemistry Research [Internet]. 2019 julio [citado 6 may 2026];58:14180-14192. Disponible en: https://doi.org/10.1021/acs.iecr.9b01456
22. Dambros JWV, Trierweiler JO, Farenzena M. Industrial Datasets and a Tool for SISO Control Loops Data Visualization and Analysis. Computers & Chemical Engineering [Internet]. 2021 marzo [citado 6 may 2026];146:107198. Disponible en: https://doi.org/10.1016/j.compchemeng.2020.107198
23. Daiguji M, Yamashita Y. An Approach for Stiction Compensation in Industrial Process Control Valves. Computers & Chemical Engineering [Internet]. 2022 febrero [citado 6 may 2026];158:107641. Disponible en: https://doi.org/10.1016/j.compchemeng.2021.107641
24. Pinto WG, Alguacil A, Bauerheim M. On the reproducibility of fully convolutional neural networks for modeling time–space evolving physical systems. Data-Centric Engineering [Internet]. 2022 abril [citado 15 mar 2026];3:e19. Disponible en: https://doi.org/10.1017/dce.2022.18
25. Alauddin M, Khan F, Imtiaz S, et al. Integrating process dynamics in data-driven models of chemical processing systems. Process Safety and Environmental Protection [Internet]. 2023 junio [citado 26 mar 2026];174:158-168. Disponible en: https://doi.org/10.1016/j.psep.2023.04.008
26. Zarzycki K, Ławryńczuk M. Long Short-Term Memory Neural Networks for Modeling Dynamical Processes and Predictive Control: A Hybrid Physics-Informed Approach. Sensors [Internet]. 2023 noviembre [citado 20 marzo 2026];23(21):8898. Disponible en: https://doi.org/10.3390/s23218898
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Víctor Germán Rodríguez Durán, Deynier Montero Gongora, Ángel Oscar Columbié Navarro, Yoalbys Retirado Mediaceja, Eddy Luis Terrero Montero, Yury Valeryevich Ilyushin, Liomnis Osorio, Alexander Mosquera Urbano
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
The journal Anales de la Academia de Ciencias de Cuba protects copyright, and operates with a Creative Commons License 4.0 (Creative Commons Attribution-NonCommercial License 4.0). By publishing in it, authors allow themselves to copy, reproduce, distribute, publicly communicate their work and generate derivative works, as long as the original author is cited and acknowledged. They do not allow, however, the use of the original work for commercial or lucrative purposes.
The authors authorize the publication of their writings, retaining the authorship rights, and assigning and transferring to the magazine all the rights protected by the intellectual property laws that govern in Cuba, which imply editing to disseminate the work.
Authors may establish additional agreements for the non-exclusive distribution of the version of the work published in the journal (for example, placing it in an institutional repository or publishing it in a book), with recognition of having been first published in this journal.
To learn more, see https://creativecommons.org
