Data‑driven numerical analysis of peristaltic Ellis nanofluid flow with motile microorganisms in a tapered channel for thermal management applications

This study investigates the peristaltic transport of a non-Newtonian Ellis nanofluid in an asymmetric tapered channel incor- porating a magnetic field, heat and mass transfer, viscous dissipation, and motile microorganisms. The model incorporates Brownian motion, thermophoresis, and porous medium effects to analyze the complex transport behavior of the fluid. The governing nonlinear equations are first expressed as a system of partial differential equations (PDEs) using the Ellis fluid model. By introducing suitable dimensionless variables, the system is converted into a set of ordinary differential equations (ODEs). Under the conditions of long wavelength and low Reynolds number, lubrication theory is then applied to simplify the governing equations. The reduced system is subsequently solved numerically in Mathematica using NDSolve. The role of key physical parameters on the velocity, temperature, nanoparticle concentration, and bioconvection characteristics is examined. The findings indicate that the Hartmann number and Ellis power-law index suppress the velocity. The temperature increases with viscous dissipation, while the nanoparticle and bioconvection concentrations decrease with increasing Schmidt and Peclet numbers, respectively. In addition, an Artificial Neural Network (ANN) framework is employed to predict the flow variables, implemented in the Python environment. The model is trained using the Adam optimization algorithm, with the hyperbolic tangent (tanh) activation function used in the hidden layers. The ANN model shows low mean squared error (MSE) and a high coefficient of determination R , indicating good agreement with the numerical solutions. The proposed model provides useful insight into magnetohydrodynamic (MHD) nanofluid transport in peristaltic flow systems relevant to biomedical transport processes and microfluidic applications.