Study on the subway environment induced by moving train using Gaussian distributed momentum source theory method

Abstract

The application of dynamic mesh method has been extensively applied for simulating piston wind-induced subway environments, e.g. air quality, which is not central processing unit-friendly in calculation as well as complex meshing required for computational fluid dynamics modelling. The feasibility of the momentum theory method was investigated to simulate moving train-induced effects on tunnel airflow and particle transport (i.e. momentum source implemented into momentum equations to simulate moving objects). A Gauss-filtered Dirac delta function was employed for Gaussian distribution of momentum source. In the current work, both momentum theory and dynamic mesh methods were employed for simulating moving train, and the reliability of the numerical simulation was validated by experimental data. The k–ε re-normalization group model was adopted for turbulence modelling. Results of the momentum theory method were reliable compared to dynamic mesh and experimental methods except during the static period. Calculation time was saved about 40% compared with dynamic mesh method. Piston wind had a strong impact on the surrounding airflow in the subway environment, leading to a further increase in the particle concentration at platforms. The momentum theory method can be efficiently applied for the simulation of moving objects in wind-induced environments.