Artificial neural network and numerical predictions on flow and heat transfer characteristics for buoyancy-driven flows in regard to dilatant fluids

The objective of present study is to carry out an unsteady numerical analysis to investigate the buoyancy-driven flows in a square enclosure containing an inner circular cylinder. The square enclosure is filled with dilatant fluids commonly known as shear thickening fluids. The inner circular cylinder is placed at the center of the square enclosure. The effects of Prandtl number (10, 100 and 1000) and Rayleigh number (10³ to 10⁶) on heat transfer characteristics are reported. The power-law index (n) varied from 1.0–1.6 where n = 1 corresponds to the Newtonian fluid. The heat transfer characteristics are almost insignificant to the change in Prandtl number. The flow and heat transfer characteristics diminished in square enclosure when filled with dilatant fluid in comparison to the Newtonian fluid. The present study demonstrates the adequacy of incorporating artificial neural network model in estimating the heat transfer characteristics.