Standard

Simulation of Flow around a Body in a Two-Dimensional Channel Using Physics-Informed Neural Networks. / Tsgoev, Ch A.; Sakharov, D. I.; Bratenkov, M. A. et al.

In: Journal of Applied Mechanics and Technical Physics, Vol. 66, No. 3, 17.12.2025, p. 487-499.

Research output: Contribution to journal › Article › peer-review

Harvard

Tsgoev, CA, Sakharov, DI, Bratenkov, MA, Travnikov, VA, Seredkin, AV, Kalinin, VA, Fomichev, DV & Mullyadzhanov, RI 2025, 'Simulation of Flow around a Body in a Two-Dimensional Channel Using Physics-Informed Neural Networks', Journal of Applied Mechanics and Technical Physics, vol. 66, no. 3, pp. 487-499. https://doi.org/10.1134/S0021894425700087

APA

Tsgoev, C. A., Sakharov, D. I., Bratenkov, M. A., Travnikov, V. A., Seredkin, A. V., Kalinin, V. A., Fomichev, D. V., & Mullyadzhanov, R. I. (2025). Simulation of Flow around a Body in a Two-Dimensional Channel Using Physics-Informed Neural Networks. Journal of Applied Mechanics and Technical Physics, 66(3), 487-499. https://doi.org/10.1134/S0021894425700087

Vancouver

Tsgoev CA, Sakharov DI, Bratenkov MA, Travnikov VA, Seredkin AV, Kalinin VA et al. Simulation of Flow around a Body in a Two-Dimensional Channel Using Physics-Informed Neural Networks. Journal of Applied Mechanics and Technical Physics. 2025 Dec 17;66(3):487-499. doi: 10.1134/S0021894425700087

Author

Tsgoev, Ch A. ; Sakharov, D. I. ; Bratenkov, M. A. et al. / Simulation of Flow around a Body in a Two-Dimensional Channel Using Physics-Informed Neural Networks. In: Journal of Applied Mechanics and Technical Physics. 2025 ; Vol. 66, No. 3. pp. 487-499.

BibTeX

@article{df20e8ae9e624433b7654cb2f244ce46,
title = "Simulation of Flow around a Body in a Two-Dimensional Channel Using Physics-Informed Neural Networks",
abstract = "This paper presents some aspects of the use of physics-informed neural networks to solve a two-dimensional stationary problem of flow around an obstacle using the Navier–Stokes equations. The influence of the activation function, quantitative parameters of the training set, adaptive regularization, and adaptive grids on the quality and accuracy of solutions is studied for a fixed neural network architecture. The relationship between these factors and modeling quality is analyzed to identify optimal conditions for increasing the accuracy and stability of solutions.",
keywords = "Navier–Stokes equations, deep learning, physics-informed neural networks, physics-informed neural networks, deep learning, Navier–Stokes equations",
author = "Tsgoev, {Ch A.} and Sakharov, {D. I.} and Bratenkov, {M. A.} and Travnikov, {V. A.} and Seredkin, {A. V.} and Kalinin, {V. A.} and Fomichev, {D. V.} and Mullyadzhanov, {R. I.}",
note = "Tsgoev, C.A., Sakharov, D.I., Bratenkov, M.A. et al. Simulation of Flow around a Body in a Two-Dimensional Channel Using Physics-Informed Neural Networks. J Appl Mech Tech Phy 66, 487–499 (2025). https://doi.org/10.1134/S0021894425700087 This work was supported by the state program “Scientific and Technological Development of Sirius Federal Territory” (agreement no. 18-03 dated September 10, 2024).",
year = "2025",
month = dec,
day = "17",
doi = "10.1134/S0021894425700087",
language = "English",
volume = "66",
pages = "487--499",
journal = "Journal of Applied Mechanics and Technical Physics",
issn = "0021-8944",
publisher = "Maik Nauka-Interperiodica Publishing",
number = "3",

}

RIS

TY - JOUR

T1 - Simulation of Flow around a Body in a Two-Dimensional Channel Using Physics-Informed Neural Networks

AU - Tsgoev, Ch A.

AU - Sakharov, D. I.

AU - Bratenkov, M. A.

AU - Travnikov, V. A.

AU - Seredkin, A. V.

AU - Kalinin, V. A.

AU - Fomichev, D. V.

AU - Mullyadzhanov, R. I.

N1 - Tsgoev, C.A., Sakharov, D.I., Bratenkov, M.A. et al. Simulation of Flow around a Body in a Two-Dimensional Channel Using Physics-Informed Neural Networks. J Appl Mech Tech Phy 66, 487–499 (2025). https://doi.org/10.1134/S0021894425700087 This work was supported by the state program “Scientific and Technological Development of Sirius Federal Territory” (agreement no. 18-03 dated September 10, 2024).

PY - 2025/12/17

Y1 - 2025/12/17

N2 - This paper presents some aspects of the use of physics-informed neural networks to solve a two-dimensional stationary problem of flow around an obstacle using the Navier–Stokes equations. The influence of the activation function, quantitative parameters of the training set, adaptive regularization, and adaptive grids on the quality and accuracy of solutions is studied for a fixed neural network architecture. The relationship between these factors and modeling quality is analyzed to identify optimal conditions for increasing the accuracy and stability of solutions.

AB - This paper presents some aspects of the use of physics-informed neural networks to solve a two-dimensional stationary problem of flow around an obstacle using the Navier–Stokes equations. The influence of the activation function, quantitative parameters of the training set, adaptive regularization, and adaptive grids on the quality and accuracy of solutions is studied for a fixed neural network architecture. The relationship between these factors and modeling quality is analyzed to identify optimal conditions for increasing the accuracy and stability of solutions.

KW - Navier–Stokes equations

KW - deep learning

KW - physics-informed neural networks

KW - physics-informed neural networks

KW - deep learning

KW - Navier–Stokes equations

UR - https://www.mendeley.com/catalogue/bc5de26c-ffe2-3ec6-9027-a89e501fd35c/

UR - https://www.scopus.com/pages/publications/105025190904

U2 - 10.1134/S0021894425700087

DO - 10.1134/S0021894425700087

M3 - Article

VL - 66

SP - 487

EP - 499

JO - Journal of Applied Mechanics and Technical Physics

JF - Journal of Applied Mechanics and Technical Physics

SN - 0021-8944

IS - 3

ER -

ID: 73717795