Artificial intelligence-based multi-expression programming prediction of magnetized radiative nanofluid flow between coaxial deformable tubes

This research primarily focuses on improving the thermal efficiency of blood between the two coaxial tubes utilizing silver nanomaterials. Endoscopy is used to detect the inner diseases of body that are complex to detect from the ordinary visualization and outer tube shows the blood flow arteries. Peristaltic waves are propagating along the wall of tube due to the contraction and expansion of outer wall. In this research novel artificial intelligence based multi expression programming approach is employed to develop a predictive model for optimizing the thermal characteristics of the fluid using numerical simulation data. The physical configuration includes a nonlinear porous inner wall of the outer tube; hence, the Darcy Forchheimer model is adopted to account for drag force effects on heat and mass transfer. Energy generation arises from fluid friction and Ohmic heating, leading to the inclusion of viscous dissipation and Joule heating in the energy equation. Thermal radiation is considered to sustain fluid isotherms, while convective boundary conditions enhance heat transport. Due to wall lubrication, slip boundary conditions are also applied. This configuration leads to complex mathematical formulations, which are solved through numerical computations using Mathematica. Further, Multi-Expression Programming (MEP) predict the thermal behaviour of flow dynamics based on numerically generated data, indicating the effectiveness of machine learning in analysing complex flow phenomena. The effect of Eckert number, Biot number, concentration of nanomaterial and Hartman number on the Nusselt number are analyzed via the predicted model of MEP. Eckert number depicts higher impact on the thermal features of fluid as compare to other parameters.