Proceedings of TUSUR University / Archive / Issue of Journal № 2, t. 28, 2025 / Neural network optimization in automated control systems for complex technological processes and production

Neural network optimization in automated control systems for complex technological processes and production

DOI: 10.21293/1818-0442-2025-28-2-153-159

Download article in PDF format

JATS xml

Abstract: The complexity and responsibility of decisions made when managing potentially hazardous and risky operations, as well as operations with a high cost of errors, often preclude the complete automation of complex technological process control. The article presents the problems of control methodology, along with the main provisions and stages of the methodology for case-based adaptation of a multi-dimensional technological process control loop to changing operating conditions. A description of the neural network adaptation procedure and principles of choosing the type and topology of the neural network adjusting the control loop of technological installations is provided. The conditions for optimal utilization of multilayer neural network architectures and formulas for calculating the number of their interneuronal connections and layers are given.

Keywords: technological process, optimization, precedent management, neural network, PI-controller

For citation:
Tyryshkin S. Yu. Neural network optimization in automated control systems for complex technological processes and production. Doklady Tomskogo gosudarstvennogo universiteta sistem upravleniya i radioelektroniki, 2025, vol. 28, no. 2, pp. 153–159. DOI: 10.21293/1818-0442-2025-28-2-153-159

Authors and copyright holders:

Tyryshkin S. Yu. , Polzunov Altai State Technical University (Barnaul, Russia)

1. Bryson A., Ho Y.-Sh. [Applied optimal control theory]. Moscow, Mir Publishers, 2012, 544 p. (in Russ.).
2. Volkova V.N., Kozlov V.N. (Eds.) Sistemnyj analiz i prinyatie reshenij: slovar'-spravochnik [System analysis and decision making: dictionary-reference book]. Moscow, Higher Education Press, 2014,614 p. (in Russ.).
3. Pikina G.A., Paschenko F.F., Paschenko A.F. O nekotoryh problemah realizacii teorii optimal'nogo upravleniya v zadachah avtomaticheskogo regulirovaniya [On some problems of implementing optimal control theory in automatic regulation tasks]. Sensors and Systems, 2022, no. 3 (262), pp. 3–8 (in Russ.).
4. Green M., Limebeer D. J. N. Linear Robust Control. NY, USA, Dover Publications, 2012, 558 p.
5. Astrom K.J., Wittenmark B. Computer-controlled systems: theory and design. Subsequent Edition. Prentice Hall, 1996, 557 p.
6. Ioannou P., Sun J. Robust adaptive control. NY, USA, Dover Publications, 2012, 852 p.
7. Markova V.D. Tsifrovaya ekonomika: uchebnik [Digital economy]. Moscow, INFRA-M, 2018, 486 p. (in Russ.).
8. Trofimov V.B., Kulakov S.M. Intellektual'nye avtomatizirovannye sistemy upravleniya tekhnologicheskimi ob"ektami [Intellectual automated control systems for technological objects]. Moscow, Infra-Inzheneriya, 2016, 232 p. (in Russ.).
9. Suratgar A.A., Tavakoli M.B., Hoseinabadi A. Modified Levenberg-Marquardt method for neural networks training. World Academy of Science, Engineering and Technology, 2005, no. 6, pp. 46–48.
10. Sousa C. Neural network learning by the LevenbergMarquardt algorithm with Bayesian regularization (part 1). Available at: http://crsouza.blogspot.com/2009/11/neural-network-learning-by-levenberg_18.html (Аccessed: May 11, 2025).
11. Bishop C.M. Pattern recognition and machine learning. Translated from English. Saint Petersburg, Dialectica, 2024, 960 p.
12. Carmona P.L., Sanchez J.S., Fred L.N.A. (Eds.) Mathematical methodologies in pattern recognition and machine learning. Contributions from the International Conference on Pattern Recognition Applications and Methods, Vilamoura, Portugal, Springer Proceedings in Mathematics & Statistics, 2012, vol. 30, 200 p.
13. Micucci U. Practical deep learning: case-based approach to understanding deep neural networks. Translated from English. Saint Petersburg, BHV-Peterburg, 2020, 368 p.
14. Agrawal Ch. Neural networks and deep learning. Translated from English. Saint Petersburg, Dialectica, 2023, 752 p.
15. LeCun Y. How machines learn: revolution in neural networks and deep learning. Translated from French. Moscow, Alpina PRO, 2021, 400 p. Available at: http://crsouza.blogspot.com/2009/11/neural-network-learning-by-levenberg_18.html (Аccessed: 11 May. 2025).