Regression analysis for thermal transport of cilia-driven MHD Sutterby liquid under joule heating and heat sink/source using ANN
R.T. Potla, Zaheer Abbas, Nouf Abdulrahman Alqahtani, Melis Arslan
Abstract
This study examines the thermal transport of cilia-driven magnetohydrodynamic flow of a non-Newtonian Sutterby fluid in a curved microchannel, considering Joule heating, heat source/sink, thermal radiation, viscous dissipation, and mixed convection. The problem is motivated by its significance in biological fluid transport and its potential applications in the development of bio-inspired microfluidic devices. The governing equations are derived in a curvilinear coordinate system with long-wavelength and low-Reynolds-number assumptions and solved numerically using the Keller Box method. To enhance prediction, an artificial neural network trained with the Levenberg-Marquardt algorithm is applied, showing excellent agreement with numerical results. Graphical and tabulated analyses are carried out to illustrate the influence of relevant physical parameters on Sutterby fluid velocities, temperature distribution, pressure gradient, and pressure rise. For validation, the findings are compared with limiting cases previously reported in the literature. Findings show that the Sutterby parameter increases velocity at the channel center, while the heat source/sink parameter strongly affects the temperature field. Findings enhance understanding of mucociliary clearance, embryo transport, and support biomedical device design, like micro-pumps, lab-on-chip, and drug delivery.