Artificial Intelligence for Modeling Lid-Driven Cavity Flow with Slip Boundary Condition

Onsoy E., Pekmen B.

7th International Conference on Intelligent and Fuzzy Systems, INFUS 2025, İstanbul, Turkey, 29 - 31 July 2025, vol.1529 LNNS, pp.225-233, (Full Text) identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 1529 LNNS
  • Doi Number: 10.1007/978-3-031-97992-7_26
  • City: İstanbul
  • Country: Turkey
  • Page Numbers: pp.225-233
  • Keywords: lid-driven cavity, machine learning, neural network, RBF method, slip boundary condition
  • TED University Affiliated: Yes

Abstract

In this study, lid-driven (LD) cavity flow with a slip condition through the moving lid is numerically and statistically studied. Firstly, the two dimensional, time dependent dimensionless governing equations are numerically resolved using the Radial Basis Function (RBF) method in space and the second order backward differentiation formula (BDF2) in time. The numerical results reveal that the resistance at the top boundary is reduced by the presence of slip parameter, and the flow features change inside the cavity. Secondly, datasets from the numerical results are collected involving Reynolds number and slip parameter as inputs, and some velocity indicators as targets. Then, Neural Network (NN) modeling for these indicators is processed. In terms of mean squared error metric, it is found that the predicted results on test data are very well suited. Instead of repeated numerical computations, NN modeling allows one to instantly determine the thermal and fluid behavior of the system at the required problem parameters.