• M. A. Novotarskyi National Technical University of Ukraine “Ihor Sikorsky Kyiv Polytechnic Institute”, Kyiv, Ukraine, Ukraine
  • V. A. Kuzmych National Technical University of Ukraine “Ihor Sikorsky Kyiv Polytechnic Institute”, Kyiv, Ukraine, Ukraine



hydrodynamic modeling, convolutional neural network, lattice Boltzmann method


Context. Solving hydrodynamic problems is associated with high computational complexity and therefore requires considerable computing resources and time. The proposed approach makes it possible to significantly reduce the time for solving such problems by applying a combination of two improved modeling methods.

Objective. The goal is to create a comprehensive hydrodynamic modeling method that requires significantly less time to determine the dynamics of the velocity field by using the modified lattice Boltzmann method and the pressure distribution by using a convolutional neural network.

Method. A method of hydrodynamic modeling is proposed, which realizes the synergistic effect arising from the combination of the improved lattice Boltzmann method and a convolutional neural network with a specially adapted structure. The essence of the method consists of implementing a sequence of iterations, each of which simulates the process of changing parameters when moving to the next time layer. Each iteration includes a predictor step and a corrector step. At the predictor step, the lattice Boltzmann method works, which allows us to obtain the field of fluid velocities in the working area at the next time layer using the field of velocities at the previous layer. At the corrector step, we apply an improved convolutional neural network trained on a previously created data set. Using a neural network allows us to determine the pressure distribution on a new time layer with a predetermined accuracy. After adding the fluid compressibility correction on the new time layer, we get a refined value of the velocity field, which can be used as initial data for applying the lattice Boltzmann method at the next iteration. Calculations stop when the specified number of iterations is reached.

Results. The operation of the proposed method was studied on the example of modeling fluid movement in a fragment of the human gastrointestinal tract. The simulation results showed that the time spent implementing the simulation process was reduced by 6–7 times while maintaining acceptable accuracy for practical tasks.

Conclusions. The proposed hydrodynamic modeling method with a convolutional neural network and the lattice Boltzmann method significantly reduces the time and computing resources required to implement the modeling process in areas with complex geometry. Further development of this method will make it possible to implement real-time hydrodynamic modeling in threedimensional domains.

Author Biographies

M. A. Novotarskyi, National Technical University of Ukraine “Ihor Sikorsky Kyiv Polytechnic Institute”, Kyiv, Ukraine

Dr. Sc., Professor of the Department of Computer Engineering

V. A. Kuzmych, National Technical University of Ukraine “Ihor Sikorsky Kyiv Polytechnic Institute”, Kyiv, Ukraine

Post-graduate student of the Department of Computer Engineering


Chen S., Doolen G. D. Lattice Boltzmann method for fluid flows, Annual Review of Fluid Mechanics, 1998, Vol. 30, pp. 329–364.

Guo Z., Shu Ch. Lattice Boltzmann Method and Its Applications in Engineering. Singapore, World Scientific, 2013, 420 p.

Tessarotto M., Fonda E., Tessarotto M. The computational complexity of traditional Lattice Boltzmann methods for incompressible fluids, Rarefied Gas Dynamics: The 26th International Symposium, Kyoto (Japan), 20–25 July 2008 : proceedings. New York, AIP Publishing, AIP Conference Proceedings, 2008, Vol. 1084, № 1, pp. 470–475,

Zimny S., Masilamani K., Jain K., Roller S. P. Lattice Biltzmann Simulations on Complex Geometries [Electronic resource]. Access mode: _Boltzmann_Simulations_on_Complex_Geometries

Korner C., Pohl T., Rude U., Thürey N., Zeiser T. Parallel Lattice Boltzmann Method for CFD Applications, Lecture Notes in Computational Science and Engineering. Berlin, Springer, 2006, Vol. 51, pp.439–466,

Xunliang L., Wen-Quan T., He Y. L. A simple method for improving the SIMPLER algorithm for numerical simulations of incompressible fluid flow and heat transfer problems, Engineering Computations. Berlin, Springer, 2005, Vol.22, pp. 921–939.

Hornik K., Stinchcombe M., White H. Multilayer feedforward networks are universal approximators, Neural Networks, 1989, Vol. 2, №5, pp. 359–366,

Lagaris I. E., Likas A. C., Papageorgiou D. G. Neuralnetwork methods for boundary value problems with irregular boundaries, IEEE Transfctions on Neural Networks, 2000, Vol. 11, №5, pp. 1041–1049,

Aarts L. P., Veer P. V. Neural network method for solving partial differential equations, Neural Processing Letters. Berlin, Springer, 2001, Vol. 14, pp. 261–271.

Smaoui N., Al-Enezi S. Modeling the dynamics of nonlinear partial differential equations using neural networks, Journal of Computational and Appliaed Mathematics, 2004, Vol. 170, pp. 27–58,

Deheuvels P., Martynov G. V. A Karhunen-Loeve decomposition of a Gaussian process generated by independent pairs of exponential random variables, Journal of Functional Analysis, 2008, Vol. 255, pp. 2363–2394

Chua L.O., Yang L. Cellular neural networks: theory, IEEE Transactions on Circuits Systems, 1988, Vol. 35, № 10, pp. 1257–1272.

Novotars'kij M. A., Nesterenko B. B. Shtuchnі nejronnі merezhі: obchis-lennya. Kyiv, Іn-t matematiki, 2004, 408 p.

He X., Luo L.-S. A Priori Derivation of the Lattice Boltzmann Equation, Physical Review E, 1997, Vol. 55, № 6, pp. R6333–R6336.

Eichinger M., Heinlein A., Klawonn A. Stationary flow predictions using convolutional neural networks, Numerical Mathematics and Advanced Applications. Berlin, Springer, 2021. – P.541–549

Koelman J.M.V.A. A simple lattice Boltzmann scheme for Navier-Stokes fluid flow, Europhysics Letters, 1991, Vol. 15, № 6, pp. 603–60.

Qian Y. H., d’Humieres D., Lallemand P.] Lattice BGK models for Navier-Stokes equation, Europhysics Letters, 1992, Vol. 17, № 6, pp. 479–484,

Sterling D. J., Chen S. Stability Analysis of Lattice Boltzmann Methods, Journal of Computational Physics, 1996, Vol. 123, № 1, pp. 196–206,

Chorin A. J. Numerical solution of the Navier-Stokes equations, Mathematics of computation, 1968, Vol. 22, № 104, pp. 745–762.

Vittoz G., Oger M., de Leffe M., Le Touze D. ] Comparisons of weakly-compressible and truly incompressible approaches for viscous flow into a high-order Cartesian-grid finite volume framework, Journal of Computational Physics, 2019, Vol. 1, pp. 100015–100048.

Kuzmych V., Novotarskyi M. Accelerating simulation of the PDE solution by the structure of the convolutional neural network modifying : Lecture Notes on Data Engineering and Communications, Artificial Intelligence and Logistics Engineering (ICAILE2022): The 2nd International Conference, 20–22 February 2022 : proceedings. Berlin, Springer, 2022, pp. 3–15.

Duchi J., Hazan E., Singer Y. Adaptive subgradient methods for online learning and stochastic optimization, Journal of Machine Learning Research, 2011, Vol. 12, pp. 2121–2159

Brandt A., McCormick S., Ruge J. Algebraic multigrid (AMG) for sparse matrix equations [Electronic resource]. Access mode: raic _multigrid_AMG_for_sparse_matrix_ equations




How to Cite

Novotarskyi, M. A., & Kuzmych, V. A. (2023). THE METHOD OF HYDRODYNAMIC MODELING USING A CONVOLUTIONAL NEURAL NETWORK . Radio Electronics, Computer Science, Control, (4), 58.



Mathematical and computer modelling