ON USE OF NEURAL NETWORKS FOR THE CONSTRUCTION OF ATTRACTION BASINS OF PERIODIC REGIMES OF NONLINEAR DYNAMIC SYSTEMS

About

News

Aims and Scope

Founder and Publisher

Editorial board

Licensing terms

Privacy Statement

Plagiarism policy

Publication ethics

Archiving Policy

Subscription

For Authors

Instructions for authors

Reviewing proccess

Copyright Notice

License agreement

Article processing charges

Archive

All issues

Search

Contacts

Contacts

V.N. Belovodskiy
Federal State Budgetary Educational Institution of Higher Education "Donetsk National Technical University", Donetsk
Research interests: modeling of technical systems, nonlinear dynamics, fractals and mathematical design, neural networks

Download pdf

UDC 519.6
DOI 10.34757/2413-7383.2023.31.4.006
Language: Russian
Annotation:The article analyzes the prospect of using neural networks for construction of attraction areas of periodic regimes of nonlinear dynamic systems. A single-mass scheme of a nonlinear vibration machine with asymmetric elastic ties and its oscillations in the subharmonic resonance zone of the order of 1:2 are considered. For the selected system parameters, two stable periodic regimes were detected in this zone. In paper, the construction of a two-layer neural network is fulfilled, its training is carried out and for certain initial points of the phase space with use of this neural network the character of the behavior of their orbits is diagnosed. The results are encouraging.
Keywords: single-mass vibrating machine, inertial vibration drive, working body, elastic ties

References:
1. Panovko, Ya.G. Introduction to the theory of mechanical vibrations: Textbook for universities [Vvedenie v teoriiu mekhanicheskikh kolebanii: Ucheb. posobie dlia vuzov] / [IA.G. Panovko ]. - Moscow: Nauka, 1991. 256 p.
2. Thompson, J.M.T. Nonlinear Dynamics and Chaos/ J.M.T.Thompson, H.B.Stewart - John Wiley & Sons, LTD, 2001, 437 p.
3. Zubov, V.I. Stability of movement (Lyapunov's methods and their application) [Ustoichivost' dvizheniia (metody Liapunova i ikh primenenie)] / [V.I. Zubov] — Moscow: Higher School, 1984. — 232 p.
4. Demidenko G.V. On the stability of solutions of quasi—linear periodic systems of differential equations [Ob ustoichivosti reshenii kvazilineinykh periodicheskikh sistem differentsial'nykh uravnenii ]/[ G.V. Demidenko, I.I. Matveeva] // Siberian Mathematical Journal. — 2004. - Vol. 45, issue 6. — pp. 1271- 1284
5. Romanov, S.A. Construction of the field of attraction based on Lyapunov functions for nonlinear systems of general form [Postroenie oblasti pritiazheniia na osnove funktsii Liapunova dlia nelineinykh sistem obshchego vida]/[ S.A. Romanov, S.E. Dushin, I.I. SHpakovskaia]. - Modern science and innovation. 2022, No. 3 (39), pp. 10-19.
6. Sokolov, D.N. Constructing estimates of areas of attraction using Lyapunov equations [Postroenie otsenok oblastei pritiazheniia s pomoshch'iu uravnenii Liapunova] [/ D.N. Sokolov] // International Journal of Humanities and Natural Sciences, vol. 4-4 (79), 2023. - pp. 190 - 195.
7. Kulizhnikov, D.B. Fields of attraction in the generalized Kapitsa problem [Oblasti pritiazheniia v obobshchennoi zadache Kapitsy]/ [D.B. Kulizhnikov, P.E. Tovstik, T.P. Tovstik // Bulletin of St. Petersburg University. Mathematics. Mechanics. Astronomy. 2019. Vol. 6 (64). Issue 3. pp. 482-492.
8. Sidorov, N.A. Areas of attraction of equilibrium points of nonlinear systems: stability, branching and destruction of solutions [Oblasti pritiazheniia tochek ravnovesiia nelineinykh sistem: ustoichivost', vetvlenie i razrushenie reshenii]/[ N. A. Sidorov, D.N. Sidorov, IU. Li ]//News of Irkutsk State University. Series "Mathematics", 2018. Vol.23. - p.46-63.
9. Hayashi, T. Nonlinear oscillations in physical systems: trans. from English [Nelineinye kolebaniia v fizicheskikh sistemakh: per. s angl. / T. KHaiasi]. — Moscow: Mir, 1968. — 432 p.
10. Smirnov, A. Some New Approaches to the Construction of Attraction Domains of Periodic Motions / A. Smirnov, V.Belovodskiy // Mechanics. Scientific Journal of Riga Technical University. Nr.36, 2015, 92-97.
11.Belovodskiy, V.N. Scanning algorithm for constructing areas of attraction of periodic modes [Skaniruiushchii algoritm postroeniia oblastei pritiazheniia periodicheskikh rezhimov] / [V.N.Belovodskii, M.IU. Sukhorukov ] // Scientific works of Donetsk National Technical University. The series "Problems of modeling and automation of dynamic systems design" (MAP-2010). Issue 8 (168): Donetsk: DonNTU, 2010. - pp.68-80
12. Ermolenko, T. V. Investigation of the effectiveness of predictive models for the system of analysis and monitoring of energy consumption at coal industry enterprises [Text][ Issledovanie effektivnosti prognosticheskikh modelei dlia sistemy analiza i monitoringa energopotrebleniia na predpriiatiiakh ugol'noi promyshlennosti [Tekst] ]/ [T. V. Ermolenko, V. N. Kotenko, V. V. Vinnik] / T. V. Ermolenko, V. N. Kotenko, V. V. Vinnik // Problems of artificial intelligence. – 2022. – № 4(27). – Pp. 25-34.
13. Antsyferov, S. S. Methodology of development of intelligent systems [Text] [Metodologiia razvitiia intellektual'nykh sistem [Tekst] ]/[ S. S. Antsyferov, A. S. Sigov, K. N. Fazilova] // Problems of artificial intelligence. – 2022. – № 2(25). – Pp. 42-47.
14. Ermolenko, T. V. Analysis of the effectiveness of deep neural network architectures for classifying images of goods [Text] [Analiz effektivnosti arkhitektur glubokikh neirosetei dlia klassifikatsii izobrazhenii tovarov [Tekst] ]/[ T. V. Ermolenko, I. E. Samorodskii ]/ T. V. Ermolenko, I. E. Samorodsky // Problems of artificial intelligence. – 2022. – № 1(24). – Pp. 54-64.
15. Ermolenko, T. V. Classification of heartbeat anomalies using deep learning [Text] [Klassifikatsiia anomalii serdtsebieniia s pomoshch'iu glubokogo obucheniia [Tekst]] /[ T. V. Ermolenko, D.V. Rolik] / T. V. Ermolenko, D.V. Roller // Problems of artificial intelligence. – 2022. – № 1(24). – Pp. 40-53.
16. Meleshchenko, N.V. Analysis of the influence of neural network parameters on the quality of time series prediction [Analiz vliianiia parametrov neironnoi seti na kachestvo prognoza vremennykh riadov /][ N.V. Meleshchenko, O.I. Fediaev] // Informatics, control systems, mathematical and computer modeling" (IUSMKM-2023): proceedings of the XIV International Scientific and Technical Conference within the IX International Scientific Forum Donetsk People's Republic/ FGBOU VO "DonNTU". – Donetsk: DonNTU, 2023, pp. 123-128.
17. Gorban, A.N. Generalized approximation theorem and computational capabilities of neural networks [Obobshchennaia approksimatsionnaia teorema i vychislitel'nye vozmozhnosti neironnykh setei]/ [A.N. Gorban' ] // Siberian Journal of Computational Mathematics, Novosibirsk: SB RAS, 1998. Vol.1, No. 1 - pp. 12-24.
18.Belovodsky, V.N. Inertial vibration machine of nonlinear type with asymmetric elastic connections [Inertsionnaia vibratsionnaia mashina nelineinogo tipa s asimmetrichnymi uprugimi sviaziami ]/ [V. N. Belovodskii, S. L. Bukin ] // Proceedings of the 6th International Conference "Modeling of nonlinear processes and systems" - Moscow: Janus - K, 2023. pp. 105-109.
19. Nikolenko, S. Deep Learning [Glubokoe obuchenie ]/[S. Nikolenko, A. Kadurin, E. Arkhangel'skaia]. - St. Petersburg: St. Petersburg, 2018. - 480 p.
20. Ketkov, Yu.L. Matlab 7: programming, numerical methods [Matlab 7: programmirovanie, chislennye metody]/[ IU.L., Ketkov, A.IU. Ketkov, M.M. SHul'ts ] - St. Petersburg: BHV-Petersburg, 2005.- 752 p.
21. Sobol, I.M. The choice of optimal parameters in problems with many criteria [Vybor optimal'nykh parametrov v zadachakh so mnogimi kriteriiami]/[ I.M. Sobol', R.B. Statnikov]. - Moscow: Nauka, 1981. - 110 p.

Issues: 4(31)'2023
Section: Math modeling
Cite: Belovodskiy, V.N. ON USE OF NEURAL NETWORKS FOR THE CONSTRUCTION OF ATTRACTION BASINS OF PERIODIC REGIMES OF NONLINEAR DYNAMIC SYSTEMS // V.N. Belovodskiy // Проблемы искусственного интеллекта. - 2023. № 4 (31). - 56-67. - http://search.rads-doi.org/project/14374/object/210542 doi: 10.34757/2413-7383.2023.31.4.006

back to issues 4(31)'2023