Thermal Radiation Effects on Fractals MHD Flow With Heat and Mass Transfer Over Rotating Porous Disk in the Presence of Dufour and Soret Using an Artificial Neural Network Approach

This paper aims to explore the new application of an intelligent numerical computational procedure based on neural networks backpropagated with the Levenberg–Marquardt scheme (NNBLMS) to investigate the thermal radiation effects on magnetohydrodynamics (MHD) flow with heat and mass transfer over a rotating porous disk in the presence of Dufour and Soret effects. The basic nonlinear coupled PDEs of thermal radiation effects on MHD flow with heat and mass transfer over a rotating porous disk in the presence of Dufour and Soret effects flow model are turned into a similar nonlinear ODE system utilizing similarity variables. A collection for NNBLMS is generated using Adam’s numerical procedure for various scenarios by varying Dufour’s number, radiation parameter, porosity parameter, Soret number, suction parameter, concentration buoyancy parameter, and Joule heating parameter. The solution of the proposed model is obtained for numerous scenarios, the NNBLMS testing, training, and validation procedures are functional, and the outcomes are associated with allusion consequences to validate the correctness of the recommended NNBLMS. The suggested NNBLMS is useful for the study and comprehension of the given flow model, as demonstrated by the error histogram, regression study, and mean square error.