LES simulation of flow field and pollutant dispersion in a street canyon under time-varying inflows with TimeVarying-SIMPLE approach

Abstract

Field observation data have shown that the actual wind speed and direction in an urban built environment always vary with time. Realizing the flow field simulation under time-varying inflows can have a great significance in fully understanding the airflow and the transportation of air pollutants in the atmospheric boundary layer. A new scheme based on the standard SIMPLE (Semi-Implicit Method for Pressure Linked Equations) algorithm has been implemented in OpenFOAM, named as TimeVarying-SIMPLE approach, to simulate the time-series airflow field under time-varying inflows in this study. The accuracy of TimeVarying-SIMPLE approach compared with PISO (Pressure-Implicit with Splitting of Operators) algorithm and PIMPLE (merged PISO-SIMPLE) algorithm was validated by solving the fully developed laminar flow in a straight pipe under sinusoidally varying inflow. The wind field and pollutant dispersion in an ideal street canyon with an aspect ratio of 1 under time-varying inflows were calculated using the TimeVarying-SIMPLE approach. The flows show an obvious flapping of the shear layer near the roof layer and the expansion or shrinkage of the main vortex under time-varying inflows, resulting in a significant increase of the removal efficiency of pollutants inside the street canyon compared to that under steady inflow. The average time of actual high-frequency wind speed data is critical to produce the time-varying inflows for the boundary conditions of numerical simulations, which could produce different simulated removal efficiency of pollutants inside the street canyon. The appropriate average time should be estimated by the pre-multiplied power spectrum of the measured high-frequency wind-speed data.