In this work we compare the DNS results (Fabregat et al. 2021, Fabregat et al. 2021) for a mild cough already reported in the literarure with those obtained with a compressible URANS equations with a k-ϵ turbulence model. In both cases, the dispersed phase has been modelled as spherical Lagrangian particles using the one-way coupling assumption. Overall, the URANS model is capable of reproducing the observed tendency of light particles under 64 µm in diameter to rise due to the action of the drag exerted by the buoyant puff generated by the cough. Both DNS and URANS found that particles above 64 µm will tend to describe parabolic trajectories under the action of gravitational forces. Grid independence analysis allows to qualify the impact of increasing mesh resolution on the particle cloud statistics as flow evolves. Results suggest that the k-ϵ model overpredicts the horizontal displacement of the particles smaller than 64 µm while the opposite occurs for the particles larger than 64 µm.
本文将现有文献中轻度咳嗽的直接数值模拟(DNS)结果与k-ϵ模型的可压缩非定常雷诺平均N-S (URANS)方程得到的结果进 行了比较. 两种情况均采用单向耦合假设, 将分散相模拟为球形拉格朗日粒子. URANS模型能够再现轻粒子(粒径小于64 µm)的上升趋 势, 这归因于咳嗽所产生浮力的阻力作用. 通过分析DNS和URANS的模拟结果发现, 在重力作用下, 大于64 µm的粒子倾向于描述抛物 线轨迹. 网格独立性分析允许随着流动的发展, 确定网格分辨率的增加对粒子云统计数据的影响. 结果表明, k-ϵ模型过度预测了直径小 于64 µm颗粒的水平位移而对于大于64 µm颗粒的预测则相反.
Keywords: Aerosol dispersion; CFD; Cough; Covid-19; DNS; Lagrangian particle tracking; URANS.
© The Chinese Society of Theoretical and Applied Mechanics and Springer-Verlag GmbH Germany, part of Springer Nature 2022.