In this study, artificial neural networks back-propagation along Levenberg-Marquardt Method is employed to predict the magnetohydrodynamics flow of penta-hybrid nanofluids with Hexanol as the base fluid over a permeable stretching surface. The governing equations for momentum and heat transfer are formulated and solved as a system of nonlinear ordinary differential equations. The results show that as convection parameter (λ) increases, the velocity profile increases while the temperature profile decreases. As porosity parameter (K) increases, the velocity profile decreases and the temperature profile increases. Increasing magnetic parameter (M) results in a decrease in the velocity profile and an increase in the temperature profile. Similarly, as heat generation parameter (δ) increases, both the velocity and temperature profiles increase. With an increase in radiation parameter (Rd), the temperature profile decreases, whereas an increase in Eckert number (Ec) causes the temperature profile to increase. As sf increases, the temperature profile decreases. Finally, increasing the volume fraction of nanoparticles leads to a decrease in the velocity profile and an increase in the temperature profile. The ANN model demonstrated its effectiveness in handling complex nonlinear relationships, providing accurate predictions with minimal error. This approach proves to be a valuable tool for analyzing temperature profiles in systems with multiple influencing factors, with potential applications in areas such as heat power generation, coating technologies, and medical devices. The model’s success underscores the potential of machine learning techniques in solving complex engineering problems. The engineering application of this work focuses on understanding and optimizing the flow behavior of magnetohydrodynamics systems, which is crucial in areas such as industrial cooling systems, heat exchangers, and energy-efficient manufacturing processes.
Ganji et al. (Sun,) studied this question.