В статье рассмотрены возможности применения подходов нейрокибернетики для осуществления интеллектуального прогнозирования риска выхода из строя компонентов судовых технических средств морских автономных надводных судов. В процессе исследования предложен алгоритм, который позволяет построить формальную модель рисков отказа наблюдаемых узлов и оборудования судовых установок на основании задач классификации и распознавания образов. Также предложен алгоритм инициализации нейронной сети, благодаря которому может быть достигнута заданная точность определения параметров работы технических средств, что дает возможность обеспечить гибкость настройки управления возникающими рисками на основании заранее определенных критериев.
риски, узлы, оборудование, судно, прогнозирование
1. Brezhnev V.l. Flexible complexes for the control of a ship power plant // Modem Science: Actual Problems of Theory and Practice. Series: Natural and technical sciences. 2020. No. 8-2. pp. 42-46.
2. Faruque, Md Omar Fuzzy Logic Based Energy Storage Management System for MVDC Power System of All Electric Ship // IEEE transactions on energy conversion. 2017. Volume 32: Number 2; pp 798-809.
3. Tang, Ruoli Optimal operation of hybrid energy system for intelligent ship: An ultrahigh-dimensional model and control method // Energy: the international journal 2020. Volume 211.
4. Epikhin, A. I.Integrated system for ensuring safe operation of technical means of ships carrying liquefied gases / A. I. Epikhin // Scientific, technical, economic and legal aspects of the development of the transport complex : Materials of the III (XVII) National Scientific and practical Conference, Novorossiysk, November 14-15, 2019. Novorossiysk: Admiral F.F. Ushakov State Maritime University, 2019. - pp. 165-167.
5. Epikhin, A. I. Building a decision support system for integrated security of gas tankers / A. I. Epikhin // Bulletin of the Admiral S.O. Makarov State University of Marine and River Fleet. - 2017. - Vol. 9. -No. 3. -pp. 481-488. -DOI 10.21821/2309-5180-2017-9-3-481-488.
6. Kondratyev, S. I. A diagnostic system of an intelligent component based on Bayesian accurate inference networks / S. I. Kondratyev, A. I. Epikhin, S. O. Malakhov // Journal of Physics: Conference Series, Novosibirsk, May 12-14, 2021. - Novosibirsk, 2021. - P. 012022. - DOI 10.1088/1742-6596/2032/1/012022.
7. Epikhin, A. I. Method for synthesis of an intelligent automatic system for diagnosing combustion engine vibration of the power supply system of an unmanned vessel / A.I. Epikhin // Journal of Physics: Conference Series, Novorossiysk, Virtual, June 15-16, 2021. - Novorossiysk, Virtual, 2021. - P. 012023. - DOI 10.1088/1742-6596/2061/1/012023.
8. Khekert, E.V. Analysis of methods to optimize control systems for power supply of marine vessels using fuzzy logic and fractal analysis / E. V. Khekert, A. I. Epikhin // Journal of Physics: Conference Series, Novorossiysk, Virtual, June 15-16, 2021. - Novorossiysk, Virtual, 2021. - P. 012088. -DOI 10.1088/1742-6596/2061/1/012088.
9. Epikhin, A. I. Neurocontrol methods in the context of development of technical solutions for transition to unmanned navigation / A. I. Epikhin // Journal of Physics: Conference Series, Novorossiysk, Virtual, June 15-16,2021. -Novorossiysk, Virtual, 2021. - P. 012115- - DOI 10.1088/1742-6596/2061/1/012115.
10. The main trends in the introduction of artificial intelligence technologies in the management of marine autonomous surface vessels / A. I. Epikhin, A.V. Ignatenko, D. E. Studenikin, E. V. Heckert// Operation of sea transport. - 2021. - № 1(98). -Pp. 88-96--DOI 10.34046/aumsuomt98/14.
11. Patent No. 2737457 Cl Russian Federation, IPC B63H21/22, G06F 17/00, G05B 17/02. Automatic system with a neuro-fuzzy network for complex technical diagnostics and control of a ship's power plant : No. 2020121876 : application 26.06.2020 : publ. 30.11.20201 A.I. Epikhin ; applicant Federal State Budgetary Educational Institution of Higher Education "Admiral F.F. Ushakov State Maritime University".
12. Epikhin, A. I. Application of neural networks based on a multilayer perceptron using fuzzy logic for technical diagnostics of ship technical means / A. I. Epikhin, S. I. Kondratiev, E. V. Heckert // Operation of sea transport. - 2020. - № 3(96). -Pp. 111-119- - DOI 10.34046/aumsuomt96/15.
13. Epikhin, A. I. Principles of neuro-management and variants of neural network architecture, applied to a complex dynamic system of SEU-VESSEL / A. I. Epikhin, E. V. Heckert, M. A. Modina // Marine intelligent technologies. - 2020. - № 4-4(50). - Pp. 18-22- - DOI 10.37220/MIT.2020.50.4.091.
14. Patent No. 2713077 Cl Russian Federation, IPC B63H 21/22, G06F 17/00, G05B 17/02. Decision support system with modular structure for operators of double-acting vessels: No. 2019114000: application 06.05.2019: publ. 03.02.2020 / A.I. Epikhin, E. V. Heckert; applicant Federal State Budgetary Educational Institution of Higher Education "Admiral F.F. Ushakov State Maritime University".
