Combined effects of particle shape, incident angle and porosity on momentum and heat transfer between spheroids and fluids

Particle-resolved direct numerical simulations are carried out to investigate the distribution of drag coefficient (Cd) and average Nusselt number (Nu) of a porous spheroid in a fluid under different Reynolds number (20 <= Re <= 200), particle shape (aspect ratio, 0.5 <= Ar <= 2.5), incident angle (0 degrees <=theta <= 90 degrees) and porosity (0.57 <=epsilon <= 0.94). Through the analysis on the numerical simulation results from 525 study cases, it is found that Cd increases first and then decreases with the increase of Ar for prolate spheroids and this trend is opposite for oblate spheroids. For both prolate and oblate spheroids, it is found that Nu decreases first and then increases with the increase of Ar. Under higher Reynolds number (100 <= Re <= 200), the effect of epsilon on Cd and Nu is noticeable. On the contrary, the effect of epsilon is almost negligible at low Reynolds number (20 <= Re < 100). Numerical results also show that when Ar < 1, Cd decreases considerably and Nu decreases slightly with the increase of theta, and this variation trend is opposite when Ar > 1. Finally, based on the numerical database, new predictive correlations of Cd and Nu for irregular porous particles in a fluid are established and the accuracy is assured by comparing the prediction results and the numerical data. These correlations can be used to improve the macroscopic multiphase models such as two-fluid model and coupled computational fluid dynamics and discrete element method. (C) 2022 Elsevier Ltd. All rights reserved.

Automated summary

This study investigates the distribution of drag coefficient (Cd) and average Nusselt number (Nu) of a porous spheroid in a fluid under different conditions using particle-resolved direct numerical simulations. The results show that the shape, Reynolds number, and porosity have noticeable effects on Cd and Nu, and there are certain trends in their variations within certain ranges. Predictive correlations are established based on the numerical data, which can be used to improve multiphase models and computational fluid dynamics methods.