Physics-informed neural network analysis of radiative MHD mixed convection in reactive nanofluid slip flow over a moving vertical plate

This study investigates radiative magnetohydrodynamic (MHD) mixed convection of a chemically reactive nanofluid with slip flow along a moving vertical plate. Conventional numerical techniques used for such strongly coupled nonlinear boundary-layer problems are often computationally intensive and less efficient for parametric analysis. To address this limitation, a Physics-Informed Neural Network (PINN) framework is employed to solve the governing equations in a mesh-free manner. The results show that magnetic effects suppress velocity, thermal radiation and viscous dissipation enhance temperature distribution, and chemical reactions reduce nanoparticle concentration. Quantitative evaluation demonstrates that the PINN achieves very low prediction errors, with MSE typically in the order of 10 − 7 – 10 − 4 , MAE below 10 − 2 , and RMSE remaining small across parameter variations, indicating stable convergence and high accuracy. The computed engineering parameters, including skin-friction coefficient, reduced Nusselt number, and reduced Sherwood number, show physically consistent trends and close agreement with benchmark numerical solutions. These findings confirm that PINNs provide an accurate and computationally efficient alternative to classical solvers and offer a reliable surrogate modeling framework for complex thermo-magnetic nanofluid transport problems relevant to energy systems and thermal engineering applications.