S-Space College of Agriculture and Life Sciences (농업생명과학대학) Dept. of Landscape Architecture and Rural System Engineering (생태조경·지역시스템공학부) Theses (Master's Degree_생태조경·지역시스템공학부)
Evaluation of Wind Pressure Coefficients of Greenhouses using Wind Tunnel Test and Numerical Model
풍동 실험과 수치 모델에 의한 온실 풍압 계수 평가
- 농업생명과학대학 생태조경·지역시스템공학부
- Issue Date
- 서울대학교 대학원
- Computational fluid dynamics; CFD validation model; Single-span Greenhouse; Wind pressure coefficient; Wind tunnel test
- 학위논문 (석사)-- 서울대학교 대학원 : 생태조경·지역시스템공학부(지역시스템공학전공), 2015. 8. 이인복.
- In the case of South Korea, greenhouse cultivation area has greatly increased with the introduction of greenhouse modernization policy by Korean government since the beginning of the 1990s. In addition, the Korean government recently announced a new development plan for a large-scale greenhouse complex to harvest higher value-added vegetables. However, many conventional greenhouses collapsed in a series of natural disasters ever year because a greenhouse facility is classified as a light-weight structure that is vulnerable to heavy wind loads. Therefore, reference to the newly modified greenhouse design standards, especially for reclaimed lands, has been required to ensure structural safety on strong winds.
In this study, to evaluate the structural safety of greenhouses according to the wind characteristics of the reclaimed lands, the wind pressure coefficients of the commercial single-span greenhouse facilities, such as Even-span, Three-quarter, Peach and Mono-span types, were measured according to various wind direction and design factors, such as the roof slope and the radius of the curvature of the roof materials. Additionally, for computing wind pressure coefficients of multi-span greenhouses, accuracy of 3D designed CFD model was initially evaluated using the measured results of Even-span and Peach type greenhouses.
Firstly, the wind environments of the reclaimed lands were designed in wind tunnel. Variations in the windward terrain roughness of the target reclaimed lands were also computed to design the wind and turbulence intensity profiles based on ESDU. Next, the wind pressure coefficients of typical single-span greenhouses in Korea were measured according to the wind direction, roof slope and radius of the curvature of the roof. From the wind tunnel measurement, when the wind blows perpendicularly to the sidewall of the greenhouse (0 and 180° wind directions), a relatively large pressure variation was generated near the eaves of the greenhouse
these large pressure differences can cause the collapse and permanent strain of the framework of the greenhouse. The wind direction on the local wind pressure was influential
therefore, consideration of the local pressure acting on the surface of the facilities is critical in establishing safe design criteria, especially glazing bars and coverings. From the results, the wind pressure coefficients of 4 types of greenhouses built in reclaimed lands were proposed in terms of structural safety and cladding design.
For CFD validations, CFD computed and WT measured results were compared with each other, and especially y+ values were mainly considered to find optimum conditions of first grid height. CFD computed y+ value almost exactly corresponded with the measured results as first grid height at 1.5 × 10-4 m. As a result, 1.5 × 10-4 m was selected for the first grid height. Computational domain and grid independence tests were also conducted to determine the domain size, the grid size. The length of upstream part was fixed at 3H, and the length of the side and the upper part was determined to be 5H and 5H, respectively. The length of downstream was determined to be 15H because the CFD model accurately predicted 10H above. The accuracy of the CFD model improved as the grid size decreased. The grid size was designed as 1.0 × 10-2 m based on a grid independence test. From a given standard, an appropriate turbulence model was selected according to the wind direction and the type and the environmental conditions of greenhouse. SST k-ω model was determined as a turbulence model for CFD validation because the statistical indices in SST k-ω model were generally higher than that in other turbulence models. Finally, the simulated and measured wind pressure coefficients were compared using statistical indices. The CFD validation model made accurate predictions under all experimental conditions. It was determined that the CFD validation model was appropriate for estimating the wind pressure coefficient.