Exploring the impact of morphological nanolayers on mixed convection in MHD nanofluids through a neurocomputational approach

Purpose

This study aims to develop a neurocomputational approach using the Levenberg–Marquardt artificial neural network (LM-ANN) to analyze flow and heat transfer characteristics in mixed convection involving radiative magnetohydrodynamic hybrid nanofluids. The focus is on the influence of morphological nanolayers at the fluid–nanoparticle interface, which significantly impacts coupled heat and mass transfer processes.

Design/methodology/approach

This research simplifies a complex system of higher-order nonlinear coupled partial differential equations governing the flow between orthogonal coaxially porous disks into ordinary differential equations via similarity transformations. These equations are solved using the shooting method, and parametric studies are conducted to observe the impact of varying important parameters. The resulting data sets are used to train, validate and test the LM-ANN model, which ensures high predictive accuracy. Machine learning and curve-fitting techniques further enhance the model’s capability to generate detailed visualizations.

Findings

The findings of this study indicate that increased nanolayer thickness (0.4–1.6) significantly improves thermal performance, while changes in the chemical reaction parameter (0.2–1) have a notable effect on enhancing the Sherwood number. These results highlight the critical role of morphological nanolayers in optimizing thermal and mass transfer efficiency in MHD nanofluids.

Originality/value

This research provides a novel neurocomputational framework for understanding the thermal and mass transfer dynamics in MHD nanofluids by incorporating the effects of interfacial nanolayers, an aspect often overlooked in conventional studies. The use of LM-ANN trained on computational data sets enables high-fidelity predictive analysis, offering new insights into the enhancement of thermal and mass transfer efficiency in hybrid nanofluid systems.