Adaptive control system of a technological production process

V. S. Smorodin ; V. A. Prokhorenko

Problemy fiziki, matematiki i tehniki, no. 3 (2021), pp. 96-102.

Résumé

The principles of construction of an adaptive control system are described for technological cycle of production with external control actions. Adaptive control feedback connections are created based on data processing for the neuroregulator model and the simulation model included in the technical means of coupling with the control system and the database for the simulation model of a technological cycle of production. Formalization of an adaptive control process that uses neural network based modeling for decision making processes is given. Neuroregulator models, their training and testing procedures are described.

Détail

Keywords: adaptive control, technological production process, neural network, neuroregulator, phase plane of states, optimal trajectory.

Bibliographie

[1] V.S. Smorodin, “Sistema upravleniya nadezhnostyu oborudovaniya veroyatnostnykh tekhnologicheskikh protsessov opasnogo proizvodstva”, Problemi programuvannya, 2007, no. 3, 107–123

[2] A.N. Goncharov, I.V. Maksimei, V.S. Smorodin, “Upravlenie rezervirovaniem i vosstanovitelnymi operatsiyami s pomoschyu imitatsionnogo modelirovaniya pri vozniknovenii otkazov v tekhnologicheskikh protsessakh opasnogo proizvodstva”, Problemy upravleniya i informatiki, 2007, no. 1, 48–60

[3] V.S. Smorodin, “Agregatnaya sistema avtomatizatsii modelirovaniya veroyatnostnykh tekhnologicheskikh protsessov proizvodstva”, Matematichni mashini i sistemi, 2007, no. 1, 105–110

[4] V.S. Smorodin, V.A. Prokhorenko, “Metod postroeniya modeli neiroregulyatora pri optimizatsii struktury upravleniya tekhnologicheskim tsiklom”, Doklady BGUIR, 2019, no. 7–8 (126), 125–132

[5] S. Osovskii, Neironnye seti dlya obrabotki informatsii, Finansy i statistika, M., 2002, 345 pp.

[6] S. Hochreiter, J. Schmidhuber, “Long shortterm memory”, Neural Computation, 9:8 (1997), 1735–1780 | DOI

[7] Y. Bengio, P. Simard, P. Frasconi, “Learning long-term dependencies with gradient descent is difficult”, IEEE Transactions on Neural Networks, 5:2 (1994), 157–166 | DOI

[8] M.T. Hagan, H.B. Demuth, “Neural networks for control”, Proceedings of the American Control Conference (San Diego, USA, 1999), v. 3, 1642–1656

[9] R.S. Sutton, A.G. Barto, Reinforcement Learning: An Introduction, The MIT Press, Cambridge, 1998

[10] J. Tsitsiklis, B.V. Roy, “An analysis of temporal-difference learning with function approximation”, IEEE Transactions on Automatic Control, 42 (1997), 674–690 | DOI | Zbl

[11] G.V. Cybenko, “Approximation by Superpositions of a Sigmoidal function”, Mathematics of Control Signals and Systems, 2:4 (1989), 303–314 | DOI | Zbl

[12] V. Mnih et al., “Human-level control through deep reinforcement learning”, Nature, 518:7540 (2015), 529–533 | DOI

[13] G. Lample, D.S. Chaplot, “Playing FPS Games with Deep Reinforcement Learning”, AAAI'17 Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence, AAAI Press, 2017, 2140–2146

[14] D. Silver et al., “Mastering the game of Go without human knowledge”, Nature, 550 (2017), 354–359 | DOI

[15] M.T. Hagan, H.B. Demuth, M.H. Beale, Neural Network Design, PWS Publishing, Boston, MA, 1996

[16] G. Klambauer, T. Unterthiner, S. Hochreiter, “Self-Normalizing Neural Networks”, Advances in Neural Information Processing Systems, 30 (2017), 972–981