O.V. Druzhinina1

1 FRС «Computer Science and Control» of RAS (Moscow, Russia)

1 ovdruzh@mail.ru

The construction and analysis of mathematical models of technical systems with switching modes of operation are relevant research problems. The range of such tasks includes the construction of generalized dynamic models, the synthesis of controls, optimal control of switchable dynamic systems under conditions of uncertainty, the development of algorithmic support using artificial neural network architectures. The aim of the work is to develop an approach to the development of algorithmic software for neural network modeling of controlled technical systems with switching. A description of a generalized multidimensional model of a switchable system is given. The problems of controlling switchable systems are characterized, which are solved using methods of numerical optimization and neural network modeling. Algorithms have been developed for modeling switchable technical systems, namely, an algorithm for finding the optimal value of a function based on a swarm of neural network agents and an algorithm for training neural network agents. The prospects of using the developed algorithmic software are considered. The results can be used in solving problems of neural network modeling, machine learning and various optimization problems, as well as in designing and improving switchable systems taking into account uncertainties. In addition, the results obtained can be used in problems of optimization of transport routes and management of transport systems.

Druzhinina O.V. Neural network algorithms for modeling switchable technical systems. Neurocomputers. 2024. V. 26. № 5. Р. 72-79. DOI: https://doi.org/10.18127/j19998554-202405-07 (In Russian)

1. Liberzon D. Switching in systems and control. New York: Springer. 2012. 233 p.
2. Li Z., Soh Y., Wen C. Switched and impulsive systems: Analysis, Design and Applications. New York: Springer. 2005. 274 p.
3. Zhao X., Kao Y., Niu B., Wu T. Control synthesis of switched systems. New York: Springer. 2017. 164 p.
4. Vasiliev S.N., Malikov A.I. On some results on the stability of switchable and hybrid systems. Actual problems of continuum mechanics. On the 20th anniversary of the IMM KAZNTS RAS. Kazan: Folio. 2011. V. 1. P. 23–81. (In Russian)
5. Bortakovskii A.S. Sufficient optimality conditions for controlling switched systems. Journal of Computer and Systems Sciences International. 2017. V. 56. № 4. P. 636–651. DOI 10.7868/S0002338817040084. (In Russian)
6. Bortakovsky A.S., Uryupin I.V. Optimization of switchable systems’ trajectories. Journal of Computer and Systems Sciences International. 2021. V. 60. № 5. P. 701–718. DOI 10.31857/S000233882105005X. (In Russian)
7. Bortakovsky A.S., Uryupin I.V. Computer technology of synthesis optimal linear switched systems. Bulletin of Computer and Information Technologies. 2019. № 11(185). P. 13–20. DOI 10.14489/vkit.2019.11.pp.013-020. (In Russian)
8. Masina O.N., Petrov A.A., Druzhinina O.V. Models for the control of technical systems motion taking into account optimality conditions. CEUR Workshop Proceedings. 2017. V. 1987. P. 386–391.
9. Druzhinina O., Masina O., Petrov A. The synthesis of the switching systems optimal parameters search algorithms. Communications in Computer and Information Science. 2019. V. 974. P. 306–320. DOI 10.1007/978-3-030-10934-9_22.
10. Haykin S. Neural Networks: A Comprehensive Foundation. Prentice Hall. 1999. 842 p.
11. Sun S., Cao Z., Zhu H., Zhao J. A Survey of Optimization Methods from a Machine Learning Perspective. IEEE Transactions on Cybernetics. 2020. V. 50. № 8. P. 3668–3681. DOI 10.1109/TCYB.2019.2950779.
12. Matrenin P.V., Sekaev V.G. System description of swarm intelligence algorithms. Software engineering. 2013. № 12. P. 39–45. (In Russian)
13. Ermakov B.S. Particle swarm optimization in the training of artificial neural networks. System analysis and logistics. 2017. № 1(14). P. 3–9. (In Russian)
14. Grachev P., Muravyov S., Filchenkov A., Shalyto A. Automata generation based on recurrent neural networks and automated cauterization selection. Information and Control Systems. 2020. № 1(104). P. 34-43. DOI 10.31799/1684-8853-2020-1-34-43. (In Russian)
15. Druzhinina O.V., Korepanov E.R., Belousov V.V., Masina O.N., Petrov A.A. Development of the instrumental support of the domestic computing platform "Elbrus 801-PC" in the problems of neural network modeling of nonlinear dynamic systems. Nonlinear World. 2021. V. 19. № 1. 2021. P. 15−28. DOI 10.18127/j20700970-202101-02. (In Russian)
16. Druzhinina O.V., Masina O.N., Petrov A.A., Lisovsky E.V., Lyudagovskaya M.A. Neural network optimization algorithms for controlled switching systems. Advances in Intelligent Systems and Computing. 2020. V. 1225. P. 470–483. DOI 10.1007/978-3-030-51971-1_39.
17. Petrov A.A., Druzhinina O.V., Masina O.N. Development of algorithmic support for modeling nonlinear control switching systems. Nonlinear World. 2022. V. 20. № 1. P. 5–13. DOI 10.18127/j20700970-202201-01. (In Russian)
18. Chernomordov S.V., Druzhinina O.V., Masina O.N., Petrov A.A. Application of machine learning methods in problems of neural network modeling of controlled technical systems. Neurocomputers. 2022. V. 24. № 1. Р. 25–35. DOI 10.18127/j19998554-202201-03. (in Russian)
19. Masina O.N., Petrov A.A., Chernomordov S.V., Openkin D.Y., Druzhinina O.V. Algorithms for switching technical systems modeling. CEUR Workshop Proceedings. 2021. V. 2922. P. 109–118.
20. Kiam H.A., Chong G., Yun L. PID control system analysis, design, and technology. IEEE Transactions on Control Systems Technology. 2005. V. 13. № 4. P. 559–576. DOI 10.1109/TCST.2005.847331.
21. Shtessel Y., Edwards C., Fridman L., Levant A. Sliding Mode Control and Observation. Basel: Birkhauser. 2014. 356 p.
22. Karpenko A.P. Modern search engine optimization algorithms. Algorithms inspired by nature: studies. Manual. Ed. 3-E. M.: Bauman Moscow State Technical University. 2021. 448 p. (In Russian)
23. Yasinsky I.F. The method of self-organization of a neural network for predicting processes with a penalty for complexity and a free structure. Bulletin of the Ivanovo State Energy University. 2013. № 3. P. 61–63. (In Russian)
24. Aggarwal C. Neural Networks and Deep Learning. Springer International Publishing. 2023. 529 p.