ANN-based analysis of thin film Maxwell fluid dynamics with electro-osmotic and nonlinear thermal effects

An intelligent Levenberg-Marquardt Technique (LMT) with artificial neural network (ANN) backpropagation (BP) has been used to analyze the thermal heat and mass transfer of unsteady magnetohydrodynamics (MHD) thin film Maxwell fluid flow in a porous inclined sheet with an emphasis on the influence of electro-osmosis. The activation energy, chemical reaction, mixed convection, melting heat, joule heating, nonlinear thermal radiation, variable thermal conductivity and thermal source/sink effect are taken into account for transport expressions. Appropriate similarity transformations were used to translate partial differential equations (PDEs) into ordinary differential equations (ODEs). After that, the built-in MATLAB BVP4C method was used for a data set assessed using the LMT-ANN strategy to solve these ODEs. The physical significance of the designed parameters is thoroughly discussed in both tabular and graphical form. The observed R -squared value is 1, and the mean square error up to 10 − 15 demonstrates the LMT-ANN's precise and accurate computing capability. The model’s validity is also confirmed by the strong agreement between the obtained predicted findings and numerical results, which shows a high degree of accuracy within the range of 10 − 8 to 10 − 11 . It was revealed that radiative heat considerably increases surface heat energy through accumulation, improving heat transfer qualities, whereas fluid temperature is raised by Joule dissipation, variable thermal conductivity, and heat source. Electro-osmosis and magnetic fields reduce fluid velocity by generating opposing forces that resist the flow. This problem works best in microscale fluid transport systems and drilling operations, where magnetic and electro-osmotic control are crucial. These systems include micro-electromechanical systems, lab-on-a-chip devices, porous geological formations, and thin film coating technologies.