Numerical Simulations of Heat Transfer Performance of Taylor–Couette Flow in Slit Model

Abstract

A partially averaged Navier–Stokes model is used to simulate the fluid field in the coaxial cylindrical gap, and the reliability of numerical simulation is verified by comparing with the particle image velocimetry experiment. Firstly, the influence of structural parameters and physical parameters on heat transfer enhancement of the slit model is investigated. In the next step, response surface method is adopted to obtain the slit parameter structure with optimal heat transfer performance. Slit width (2.5 mm < w < 15 mm), slit number (9 < N < 15), Reynolds number (2000 < Re < 4652) and Prandtl number (5.90 < Pr < 6.22) are selected as design parameters while the average Nusselt number is taken as the objective function. The results show that the increasing of Reynolds number strengthens the jet flow of vortex pairs, which enhances the heat transfer capacity of the Taylor vortex. As increasing the slit width, heat transfer performance of the model increases first and then decreases. The optimized model with slit structure parameters of N = 12 and w = 13.15 mm has the best heat transfer capacity, which increases by 12.42% when Reynolds Numbers is 4652.