Three-dimensional numerical simulation of a jet flow around an underwater vehicle in a pool intended to collect training data for a neural network

Three-dimensional numerical simulation of a jet flow around an underwater vehicle in a pool intended to collect training data for a neural network

Nikolay A. Tschur
PhD in Physics and Mathematics, Russian State Scientific Center for Robotics and Technical Cybernetics (RTC), Mathematician, 21, Tikhoretsky pr., Saint Petersburg, 194064, Russia, This email address is being protected from spambots. You need JavaScript enabled to view it., RecearcherID: AAH-8421-2019, ORCID: 0000-0003-2480-6861

Aleksei A. Pozhilov
RTC, Engineer, 21, Tikhoretsky pr., Saint Petersburg, 194064, Russia, This email address is being protected from spambots. You need JavaScript enabled to view it., ORCID: 0009-0005-2458-7533

Received December 23, 2024

Abstract
Nowadays, creation of control systems with use of neural networks is becoming increasingly relevant and promising. In order to collect training data for the neural network, a parameterized computational model for a series of three-dimensional numerical simulations of a jet flow around an underwater vehicle in a pool was built in present work. The input parameters for the parametrized computational model are the position of the jet source and the device in the pool, and the output parameters are the forces and moments acting on the device. Using the created computational model, 3 series of calculations were performed, in each of series results were obtained for 18 different positions of the device.

Key words
Computational fluid dynamics (CFD), jet, neural network, training data, remotely operated underwater vehicle (ROUV), autonomous underwater vehicle (AUV).

Acknowledgements
The work was carried out as the part of the state task of the Russian Ministry of Education and Science «Investigation of methods of using neural networks for modeling nonlinear nonstationary physical fields in anisotropic media in solving problems of fracture mechanics and fluid dynamics of turbulent flows» (FNRG-2024-0001 1023101000004-2-1.2.1;2.3.1 №075-00697-24-03 from 04.07.2024).

Bibliographic description
Tschur, N.A. and Pozhilov, A.A. (2025), "Three-dimensional numerical simulation of a jet flow around an underwater vehicle in a pool intended to collect training data for a neural network", Robotics and Technical Cybernetics, vol. 13, no. 2, pp. 115-120, EDN: EEIMXC. (in Russian).

EDN
EEIMXC

UDC identifier
629.58:004.032.26

References

  1. Zhao, J. (2018), “Neural Network Predictive Control for Autonomous Underwater Vehicle with Input Delay”, Journal of Control Science and Engineering,2018, pp. 1-8, DOI: 10.1155/2018/2316957.
  2. Eski I. and Yildimir S. (2014), “Design of Neural Network Control System for Controlling Trajectory of Autonomous Underwater Vehicles”, International Journal of Advanced Robotic Systems,11, no. 7, pp. 1-17, DOI: 10.5772/56740.
  3. Forouzantabar A., Gholami, B. and Azadi, M. (2012), “Adaptive Neural Network Control of Autonomous Underwater Vehicles”, International Journal of Electrical and Computer Engineerin,6, no. 7, pp. 866-871.