Hybrid neuro-fuzzy and artificial neural network modeling of radiative MHD sisko nanofluid flow over stretching and shrinking surfaces with joule heating and viscous dissipation

A coupled numerical and intelligent modeling strategy is developed to analyze radiative magnetohydrodynamic flow of a non-Newtonian Sisko nanofluid over stretching and shrinking surfaces in the presence of Joule heating, nonlinear viscous dissipation, Brownian diffusion, thermophoresis, and motile microorganism transport. The mathematical formulation consists of boundary-layer equations governing momentum, thermal energy, solutal concentration, and microorganism density, incorporating Lorentz force effects and nonlinear shear-dependent viscosity. By employing suitable similarity transformations, the governing partial differential equations are reduced to a highly nonlinear system of ordinary differential equations and solved numerically using a boundary value approach. The numerical solutions reveal the existence of dual solution branches for shrinking surfaces, indicating bifurcation phenomena governed by viscous dissipation and stretching/shrinking intensity. The computed datasets are subsequently utilized to construct Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) surrogate models for accurate prediction of skin friction, heat transfer, mass transfer, and microorganism density rates. A close correspondence between numerical and intelligent predictions is observed, with correlation coefficients exceeding ۰.۹۹. The analysis demonstrates that Joule heating and viscous dissipation markedly elevate the thermal field while diminishing solutal and nanoparticle concentration levels, whereas magnetic interaction suppresses fluid motion and thickens the thermal boundary layer.

کلیدواژه ها:

Neural Network ، Sisko nanofluid ، Stretching/Shrinking Surface ، thermal radiation ، Viscous Dissipation