Unsupervised AI-driven numerical study of flow and heat transfer analysis of nano-encapsulated phase change material nanofluid over wedge

This study deals with the analysis of three dimensional magnetohydrodynamic (MHD) flow of a nano-encapsulated phase change materials (NEPCMs) based nanoliquid over a wedge. Here, buoyancy driven flow with Hall and ion-slip effects are accentuated and the flow is examined under velocity and temperature slips. The energy equation is based on Fourier’s law of heat conduction, accounting the Ohmic heating and thermal radiation as the key processes in heat transfer. The governing non-linear partial differential equations of the flow and energy dynamics are reduced to a set of non-linear ordinary differential equations using similarity transformations. Physics informed neural networks (PINNs) are the primary method used to solve the resulting system which incorporates the physical laws and boundary together with the structure of the network. This study shows that an increase in Hall and ion parameters lowers skin friction coefficient and increases rate of heat transfer. The present work validates the effectiveness of PINNs in solving three-dimensional flow problems that may offer crucial information about the steady MHD nanofluid behaviour in NEPCM systems. These findings are significant to improve thermal transport and energy system concepts.