Design of backpropagation learning algorithm for MHD mixed convective Prandtl nanofluid flow with activation energy

Alshehery, Eman Fayz; Alaidarous, Eman Salem; Alharbey, Rania A.; Zahoor Raja, Muhammad A

doi:10.21608/djs.2024.271404.1152

Design of backpropagation learning algorithm for MHD mixed convective Prandtl nanofluid flow with activation energy

Document Type : Review Article

Authors

, Faculty of Science, King Abdul Aziz University

10.21608/djs.2024.271404.1152

Abstract

The use of artificial intelligence techniques for solving challenges has grown in popularity recently in a range of areas. Additionally, nanofluid is interesting for a variety of applications, especially in cooling and heat transfer systems, since it is used to improve the thermal features of fluid. In the present study, a design of a backpropagation learning algorithm is provided to analyze the flow properties in a magnetohydrodynamic mixed convective flow of Prandtl nanofluid (MHD-MCPNFF) with gyrotactic microorganisms over a stretchable surface affected by the activation energy. An ordinary differential equations ODEs system is obtained from a partial differential equations PDEs system of the original mathematical formulation by using suitable transformations. Applying the Lobatto IIIA technique to solve ODEs for various scenarios by changing the values of Prandtl fluid parameter (α), magnetic parameter (M), Brownian motion (Nb), thermophoresis (Nt), activation energy (E), chemical reaction rate (σ), and Peclet number (Pe) to find a set of data for the MHD-MCPNFF model. Using these solutions through nftool in MATLAB for designing the Levenberg–Marquardt backpropagation learning algorithm (LMBLA). The effectiveness and accuracy of the designed LMBLA are verified through the mean squared error (MSE), error histograms, and regression illustration plots. The flow velocity has the opposite behavior for growing values for Prandtl fluid and magnetic parameters. For rising values of Brownian motion and thermophoresis parameters, the fluid temperature increases. The increasing values of the activation energy parameter imply the increasing concentration of nanoparticles.

Keywords