Analysis of MHD Williamson hybrid nanofluid flow with Cattaneo–Christov heat flux over a stretching cylinder: An artificial intelligence approach

Source abstract

Abstract This study presents an artificial intelligence–assisted numerical investigation of magnetohydrodynamic (MHD) Williamson hybrid nanofluid (HNF) flow with Cattaneo–Christov heat and mass flux over a stretching cylinder embedded in a porous medium. The base fluid, modeled as blood, is loaded with copper oxide (CuO) and copper (Cu) nanoparticles to enhance thermal performance. The governing nonlinear boundary‐layer equations are formulated using similarity transformations and solved using MATLAB's bvp4c solver. The obtained solutions are further refined using a Levenberg–Marquardt‐based artificial neural network (LM‐ANN) to achieve improved predictive accuracy and computational efficiency. The influence of key parameters—including magnetic field strength, the Williamson fluid parameter, nanoparticle volume fraction, thermal relaxation time, and chemical reaction rate, on velocity, temperature, and concentration distributions is systematically analyzed. The results show that increasing the Williamson parameter enhances fluid velocity while reducing thermal and concentration boundary layer thicknesses, whereas the magnetic field produces the opposite effect. Hybrid nanoparticles substantially improve heat transfer rates, highlighting their potential in biomedical and engineering applications such as targeted drug delivery, thermal therapy, and micro‐scale heat exchangers. The proposed AI‐augmented methodology not only accelerates computation but also offers robust generalization capabilities for predicting flow behavior under varying conditions, providing a valuable framework for HNF transport modeling in porous media.