CFD simulations can be adequate for the evaluation of snow effects on structures
CFD simulations can be adequate for the evaluation of snow effects on structures
复制标题
DOI：
10.1007/s12273-020-0643-0
复制
发表时间：
2020-08
影响因子：
5.5
通讯作者：
Y. Tominaga;T. Stathopoulos
中科院分区：
文献类型：
--
作者：
Y. Tominaga;T. Stathopoulos
在特定的降雪和风力条件下，由于建筑物几何形状的不同，雪粒运动与流体流动之间的复杂相互作用会在建筑物屋顶上形成大量的雪堆。特别是对于跨度大、形状复杂的屋顶，屋顶上的飘雪行为和由此产生的堆积与地面上的完全不同。不平衡的雪荷载和由于雪堆引起的屋顶上的悬雪可能导致建筑物或其部件或组件的开裂甚至倒塌（Zallen 1988; Peraza 2000; O 'Rourke 2008）。因此，一些协会和组织已采用建筑规范/标准规定，以便在结构设计过程中考虑雪荷载（AIJ 2019; ASCE 2017; ISO 2013等）。这些规范规定基于过去的观察和实验，以反映结构荷载设计上预期的不平衡雪荷载。然而，在各种天气和建筑条件下，包括位于主体屋顶附近的其他建筑物的影响，很难实施这些规范规定（Flaga等人，2019）。因此，有必要找到另一种方法来预测，高精度，屋顶雪荷载的空间分布在特定的建筑物（欧文1997）。传统上，使用人造颗粒模拟雪的风洞和水洞实验已用于此类目的（Isyumov和Davenport 1974; Anno和Konishi 1981;欧文和威廉姆斯1983; Anno和Tomabechi 1985; Anno等人1986; Zhou等人2016 a，B; Flaga和Flaga 2019; Flaga等人2019）。然而，这种设施并不总是可用的，而且通常昂贵和耗时。此外，它们在相似律方面有严重的局限性（Kind 1976，1986; Iversen 1981; Anno 1984; Peterka and Petersen 1990）。近几十年来，计算流体动力学（CFD）技术已被积极用于建筑物周围的环境风工程问题，例如行人风，通风和分散（Blocken 2014）。在CFD模拟中，需要做出大量的选择。物理模型、边界条件和数值参数应适当，因为这些选择可能对结果产生重大影响。因此，在建筑物周围流动的计算流体力学的使用中，已经提出了一些最佳实践指南，例如Franke等人的指南。(2007，2011）、Britter和Schatzmann（2007）、Tominaga et al.（2008 b）、Blocken and瓜尔蒂耶里（2012）和Blocken（2015）。最近的研究表明，CFD是相当成功的环境风工程问题，至少当高质量和高分辨率的网格与适当的边界条件的应用。CFD的繁荣主要归功于对平均风速评估的高置信度，这些平均风速构成了一些非常有用的信息，并且在环境风工程应用中几乎总是准确的。特别是，近年来，使用CFD研究建筑物周围的污染物扩散迅速增加，并且表现出普遍良好的性能（Tominaga和Stathopoulos 2013，2016）。在雪颗粒的三种输运过程中，即蠕动、跃移和悬浮（Bagnold 1941），污染物扩散的机理可认为与雪颗粒悬浮的机理相似。剩下的考虑是将重力效应
Under specific conditions of snowfall and wind, a large amount of snowdrift forms on building roofs as the result of the complex interaction between snow particle motion and fluid flow due to building geometry. In particular, for roofs with large spans and complex shapes, drifting snow behavior and resulting accumulation on the roof are completely different from those on the ground. Unbalanced snow loads and overhanging snow on roofs due to snowdrifts may lead to the cracking or even a collapse of buildings or their parts or components (Zallen 1988; Peraza 2000; O’Rourke 2008). Therefore, building code/standard provisions have been adopted by several societies and organizations in order to consider the snow load in structural design processes (AIJ 2019; ASCE 2017; ISO 2013, etc.). These code provisions are based on past observations and experiments to reflect the expected unbalanced snow loads on the structural load design. However, it is difficult to implement these code provisions under various weather and building conditions including the influence of other buildings located near the subject roof (Flaga et al. 2019). Therefore, it is necessary to find another approach to predict, with high accuracy, the spatial distributions of roof snow loads around specific buildings (Irwin 1997). Traditionally, wind and water tunnel experiments using artificial particles imitating snow have been used for such purposes (Isyumov and Davenport 1974; Anno and Konishi 1981; Irwin and Williams 1983; Anno and Tomabechi 1985; Anno et al. 1986; Zhou et al. 2016a, b; Flaga and Flaga 2019; Flaga et al. 2019). However, such facilities are not always available and are usually expensive and time consuming. Furthermore, they have serious limitations with regard to the similarity law (Kind 1976, 1986; Iversen 1981; Anno 1984; Peterka and Petersen 1990). In recent decades, the Computational Fluid Dynamics (CFD) technique has been actively used for environmental wind engineering problems around buildings, eg, pedestrian winds, ventilation, and dispersion (Blocken 2014). In CFD simulations, a large number of choices need to be made. The physical models, boundary conditions, and numerical parameters should be appropriate, given that these choices may have a significant impact on the results. Consequently, several best practice guidelines in the use of CFD for flow around buildings have been proposed, such as those by Franke et al.(2007, 2011), Britter and Schatzmann (2007), Tominaga et al.(2008b), Blocken and Gualtieri (2012), and Blocken (2015). Recent studies show that CFD is rather successful for environmental wind engineering problems, at least when high-quality and high-resolution grids with appropriate boundary conditions are applied. The prosperity of CFD is mainly attributed to the high level of confidence on the evaluation of mean wind velocities, which constitute some very useful information and almost always accurate in environmental wind engineering applications. In particular, the use of CFD to study pollutant dispersion around buildings has increased rapidly in the recent years and has shown generally good performance (Tominaga and Stathopoulos 2013, 2016). Among the three transport processes of snow particles, ie, creep, saltation and suspension (Bagnold 1941), the mechanism of pollutant dispersion may be considered similar to that of snow particle suspension. The remaining considerations are to incorporate the gravitational effect of