Yu Hesheng, Thé Jesse
a Lakes Environmental Research Inc. , Waterloo , Ontario , Canada.
b Department of Mechanical and Mechatronics Engineering , University of Waterloo , Waterloo , Ontario , Canada.
J Air Waste Manag Assoc. 2017 May;67(5):517-536. doi: 10.1080/10962247.2016.1232667. Epub 2016 Sep 20.
The dispersion of gaseous pollutant around buildings is complex due to complex turbulence features such as flow detachment and zones of high shear. Computational fluid dynamics (CFD) models are one of the most promising tools to describe the pollutant distribution in the near field of buildings. Reynolds-averaged Navier-Stokes (RANS) models are the most commonly used CFD techniques to address turbulence transport of the pollutant. This research work studies the use of [Formula: see text] closure model for the gas dispersion around a building by fully resolving the viscous sublayer for the first time. The performance of standard [Formula: see text] model is also included for comparison, along with results of an extensively validated Gaussian dispersion model, the U.S. Environmental Protection Agency (EPA) AERMOD (American Meteorological Society/U.S. Environmental Protection Agency Regulatory Model). This study's CFD models apply the standard [Formula: see text] and the [Formula: see text] turbulence models to obtain wind flow field. A passive concentration transport equation is then calculated based on the resolved flow field to simulate the distribution of pollutant concentrations. The resultant simulation of both wind flow and concentration fields are validated rigorously by extensive data using multiple validation metrics. The wind flow field can be acceptably modeled by the [Formula: see text] model. However, the [Formula: see text] model fails to simulate the gas dispersion. The [Formula: see text] model outperforms [Formula: see text] in both flow and dispersion simulations, with higher hit rates for dimensionless velocity components and higher "factor of 2" of observations (FAC2) for normalized concentration. All these validation metrics of [Formula: see text] model pass the quality assurance criteria recommended by The Association of German Engineers (Verein Deutscher Ingenieure, VDI) guideline. Furthermore, these metrics are better than or the same as those in the literature. Comparison between the performances of [Formula: see text] and AERMOD shows that the CFD simulation is superior to Gaussian-type model for pollutant dispersion in the near wake of obstacles. AERMOD can perform as a screening tool for near-field gas dispersion due to its expeditious calculation and the ability to handle complicated cases. The utilization of [Formula: see text] to simulate gaseous pollutant dispersion around an isolated building is appropriate and is expected to be suitable for complex urban environment.
Multiple validation metrics of [Formula: see text] turbulence model in CFD quantitatively indicated that this turbulence model was appropriate for the simulation of gas dispersion around buildings. CFD is, therefore, an attractive alternative to wind tunnel for modeling gas dispersion in urban environment due to its excellent performance, and lower cost.
由于流动分离和高剪切区域等复杂的湍流特征，气态污染物在建筑物周围的扩散十分复杂。计算流体动力学（CFD）模型是描述建筑物近场污染物分布最有前景的工具之一。雷诺平均纳维-斯托克斯（RANS）模型是处理污染物湍流输运最常用的CFD技术。本研究工作首次通过完全解析粘性子层，研究了[公式：见正文]封闭模型在建筑物周围气体扩散中的应用。同时还纳入了标准[公式：见正文]模型的性能进行比较，以及一个经过广泛验证的高斯扩散模型——美国环境保护局（EPA）的AERMOD（美国气象学会/美国环境保护局监管模型）的结果。本研究的CFD模型应用标准[公式：见正文]和[公式：见正文]湍流模型来获得风流场。然后基于解析得到的流场计算被动浓度输运方程，以模拟污染物浓度分布。风流场和浓度场的模拟结果通过大量数据，采用多种验证指标进行了严格验证。[公式：见正文]模型能够较好地模拟风流场。然而，[公式：见正文]模型无法模拟气体扩散。在流动和扩散模拟方面，[公式：见正文]模型均优于[公式：见正文]模型，无量纲速度分量的命中率更高，归一化浓度的“2倍因子”（FAC2）更高。[公式：见正文]模型的所有这些验证指标均通过了德国工程师协会（Verein Deutscher Ingenieure，VDI）指南推荐的质量保证标准。此外，这些指标优于或等同于文献中的指标。[公式：见正文]模型与AERMOD性能的比较表明，在障碍物近尾流中的污染物扩散方面，CFD模拟优于高斯型模型。由于其计算速度快且能够处理复杂情况，AERMOD可作为近场气体扩散的筛选工具。利用[公式：见正文]模型模拟孤立建筑物周围的气态污染物扩散是合适的，并且有望适用于复杂的城市环境。
CFD中[公式：见正文]湍流模型的多种验证指标定量表明，该湍流模型适用于模拟建筑物周围的气体扩散。因此，由于其出色的性能和较低的成本，CFD是城市环境中气体扩散建模的一种有吸引力的替代风洞的方法。
