Machine-Learning Modelling of a Fluid Flow in a Two-Dimensional Channel

Over the past decade, the integration of Artificial Neural Networks (ANNs) has garnered significant interest, capitalizing on their ability to discern intricate patterns within data. Focused on enhancing computational efficiency, this article explores the application of ANNs in fluid-dynamics simulations, particularly in modelling a benchmark problem of fluid flow in a two-dimensional (2D) channel. Leveraging a Multilayer Perceptron trained on Finite Volume Method numerical data, for both interpolation and extrapolation estimations and various grid resolutions, our findings demonstrate the ANN's prowess as a swift and accurate surrogate for traditional numerical methods. Overall, the results of this work mark a pioneering step towards leveraging Machine Learning for modelling complex relationships in fluids phenomena, promising transformative advancements in Computational Fluid Dynamics.