Journal Article

ANN-based prediction of conjugate convective flow of micropolar nanofluids in inclined porous enclosures with Lorentz force

Ahmed, Sameh E., Thumma, Thirupathi, Mansour, M. A., Hafed, Zahra, Alhazmi, Muflih, Mishra, S. R., Mohamed, Eman F., Bakier, M. A. Y.

2025

Overview

Efficient heat transfer in inclined enclosures is critical for applications in thermal management, energy storage, and electronic cooling, yet the combined effects of micropolar nanofluids, porous media, and electromagnetic forces remain underexplored. This study investigates conjugate convective heat transfer in a porous inclined cavity filled with micropolar nanofluid under a tilted Lorentz force, where local thermal non-equilibrium is assumed between fluid and solid phases. The governing nonlinear equations are solved using the finite difference method (FDM), with adiabatic vertical walls and thermally conductive horizontal walls. To reduce computational cost, an artificial neural network (ANN) is trained on FDM-generated data to predict local Nusselt numbers. The results show that increasing the thickness of the solid wall from 0.05 to 0.3 reduces the maximum temperature by up to 84.28%, indicating improved thermal insulation characteristics. Additionally, higher solid volume fractions (up to 0.2) and stronger micropolar effects (vortex viscosity ratio up to 2.0) increase thermal resistance, resulting in a reduction in heat transfer of approximately 20%. Furthermore, enhancing the porosity of the medium from 0.1 to 0.9 leads to a 76.67% improvement in convective flow. This work advances the state of the art by coupling micropolar nanofluid dynamics, porous media, and tilted magnetic fields in inclined enclosures—an area not previously addressed with such detail. The integration of ANN with physics-based modeling offers a novel, high-fidelity, and computationally efficient framework for the optimization of complex thermal systems.

• Publisher: Springer Berlin Heidelberg,Springer Nature B.V
• Subject: Artificial intelligence / Artificial neural networks / Boundary conditions / Brownian motion / Computational efficiency / Conjugates / Convective flow / Convective heat transfer / Cooling / Design optimization / Efficiency / Electromagnetic forces / Enclosures / Engineering / Equilibrium / Finite difference method / Fluid flow / Geometry / Heat conductivity / Heat transfer / Influence / Investigations / Lorentz force / Machine learning / Magnetic fields / Mechanical Engineering / Nanofluids / Neural networks / Nonlinear equations / Nusselt number / Porous media / Radiation / Rayleigh number / Reynolds number / Solid phases / Technical Paper / Temperature / Thermal insulation / Thermal management / Thermal resistance / Viscosity / Viscosity ratio