RANS CFD simulations can be successfully used for simulating snowdrift on roofs in a long period of snowstorm
- PDF / 1,322,582 Bytes
- 7 Pages / 612 x 808 pts Page_size
- 10 Downloads / 191 Views
Position Paper
RANS CFD simulations can be successfully used for simulating snowdrift on roofs in a long period of snowstorm Xuanyi Zhou (), Yu Zhang, Luyang Kang, Ming Gu State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, Shanghai 200092, China
Article History Received: 01 September 2019 / Revised: 02 April 2020 / Accepted: 15 April 2020 © Tsinghua University Press and Springer-Verlag GmbH Germany, part of Springer Nature 2020
1
Introduction
A long period of snowstorm, namely a snowdrift event, could last for hours to several days and the snow distribution on roof varies significantly during this period. Thus, for the heavy-snowfall regions, it is important to consider the snowdrift when evaluating the snow load on roof structures. Snowdrift on roofs can cause many problems, such as unbalanced snow loads and excessive partial loads (e.g. Zallen 1988; Tsuchiya et al. 2002). At the worst, a significant amount of snowdrift will lead to the collapse or cracking of building (O’Rourke 2008). Field measurements are the most reliable approach to study the snowdrift. However, the environmental condition is neither controllable nor repeatable in the field measurements, the applicability of the on-site outdoor measurements is limited for systematic studies (Tominaga 2018). In contrast to field measurement, wind tunnels or water flumes can provide a controllable platform to study snowdrift on roof surfaces. However, experiments inevitably suffer from the problem of similarity requirements (e.g. Blocken 2014; Zhou et al. 2014, 2016b, 2016c, 2019). In many previous studies, notably Isyumov (1971), Iversen (1979, 1980, 1981), Kind (1976, 1986), Kind and Murray (1982), Anno (1984), a number of similarity requirements were proposed to reasonably simulate snowdrift around structures or on building roofs, but they are difficult to be all matched in reduced-scale experiments. Therefore, for systematic and parametric studies, the computational fluid dynamics (CFD) technique is expected to be a powerful tool for addressing the problems of snowdrift (Tominaga 2018). CFD simulations can provide detailed information on the relevant flow variables in the entire calculation domain E-mail: [email protected]
under well-controlled conditions and without similarity constraints (e.g. Blocken 2014, 2015; Tominaga 2018; Zhou et al. 2019). In recent decades, with the dramatic increase in computational power, CFD technique is frequently used by researchers to simulate the snowdrift on roofs. Many CFD approaches exist, including Reynolds-averaged NavierStokes simulations (RANS), unsteady RANS (URANS), large eddy simulation (LES), and hybrid LES-RANS, etc. In the field of simulating snowdrift on roofs, only RANS (including RANS and URANS) were employed in the former literatures. The basic problem in snowdrift simulation is how to obtain the reliable flow field since snowdrift on roofs is closely associated with the airflow patterns, especially the shear stress, near the snow surface. RANS offers the most ec
Data Loading...
