No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Wind effect on scallop domes with negative amplitude and prominence using Experimental and Numerical Study
Abstract
This study investigates the wind effect on hexa-sectored scallop domes using Computational Fluid Dynamics (CFD) and wind tunnel tests. Similar to the shape of a seashell, the scallop dome is one of the most common domes used to cover large spans. The scallop dome has an additional curvature equal to its sectors compared to a base dome (spherical). Hence, it has better structural efficiency compared to a spherical dome under the same condition. The curvatures on the scallop dome create alternate “ridges” and “grooves” on it. According to the computations of the article, the grooves created on this dome in the sector, as well as their position angle to the wind direction significantly affect pressure coefficient value. Due to these differences, wind pressure on scallop domes significantly differs from wind pressure on spherical domes. The results indicate that the ridges cause negative wind pressure coefficients whose magnitude reaches a maximum of −2.0 for an angle of alignment of 30°. The analyses have been conducted through Ansys-Fluent. This study presents the equation of wind pressure coefficient in the most critical of dome groove positions. The arch created between the grooves of a scallop dome is another effective parameter on maximum negative pressure. In the case study scallop dome, if the wind effect angle is considered α = 15, the maximum deformation in the structure will be created, which is 20% higher than that of α = 0.
... Khosrowjerdi Analyzes were performed through ANSYS Fluent software. The results showed that the wind pressure on scallop domes is significantly different from the wind pressure on spherical domes [28]. ...
Climatic conditions have a great impact on the erosion of the coverage and the materials destruction of the dome gradually. Therefore, studying the shape and form of the dome in historic mosques can greatly assist to identify the affected points of the different types of the domes and provide solutions to prevent early destruction of the domes. In the present work, the turbulent flow of wind around the four samples of different domes is investigated, using ANSYS CFX software, to determine which parts of the dome geometry are most affected by wind and erosion. In the present work, for the first time, it will be tried to study different types of domes used in different climates and their geometric shapes, besides conducting research to prevent early erosion. The results demonstrated that a large vortex has shaped on the opposite side of the wind, which affects the area behind the dome and causes a negative pressure through velocity reduction. Also, the highest wind velocity is formed a little higher and hinder of the dome. The results of the shear stress on the crown of the dome for the four cases illustrated that for the dome type W4, the highest shear stress is about 15 Pa on the face against the wind and it is about 12 Pa for that of W1 on the face opposite the wind. It should be noted that the position of the most stresses on the dome crown corresponding to the most damage to the building is estimated.
Pyramidal domes are a type of roof covering that can have various numbers of faces. Since most of these domes are constructed from lightweight structures, such as space frames, accurately calculating the wind load the primary lateral load for this structure type is essential. Due to the absence of wind pressure coefficients for this structure type in standard design codes, either numerical modeling or wind tunnel testing must be used to calculate wind loads. This study explores wind tunnel models at four height-to-span ratios (0.25, 0.5, 0.75, and 1) for six-sided and eight-sided domes. Additionally, 6-, 8-, and 12-sided domes are modeled using ANSYS based on the computational fluid dynamics (CFD) method. Results show that, when the protrusions of the sectors are positioned at angles between 60° and 90°, negative pressure coefficients increase locally in these areas. Furthermore, the study presents equations for the pressure coefficients of these domes through modeling of conical domes.
The geometry of a building is one of the influential parameters in calculating the wind load. This parameter is denoted as Cp in static methods. It is provided in codes for conventional buildings. The present study calculates the wind pressure coefficient for a non-conventional Y-shaped building model with a scale of 1:500. To obtain the wind effects on the building, the wind tunnel test and computational fluid dynamics (CFD) were employed. The model was rotated in 10-degree rotation steps within the wind tunnel to find the most critical wind load orientation. Also, the wind pressure was applied to the building in the CFD approach to obtain the maximum displacement. The largest displacement (relative to the original position at θ = 0°) was found to be 3.75 at θ = 30°. The numerical results were found to be in good agreement with the experimental wind tunnel results. This demonstrated that the proposed method could be a suitable alternative to the wind tunnel test.
Large-eddy simulations were used to investigate unsteady flows around a wall-mounted hemisphere as the Reynolds number (Re, based on the diameter of the hemisphere D) increased from 7 × 10⁴ to 7 × 10⁵. The hemisphere was immersed in a low-turbulence-intensity boundary layer with a thickness of δ/D = 0.5. Strong Re dependence was confirmed to be present even for the flow around a wall-mounted obstacle after systematic examination of aerodynamic forces, local pressures, and flow structures. Drag and lift crises were observed simultaneously, with the critical Re noted at approximately 3 × 10⁵. As with circular cylinders and spheres, a laminar-turbulent transition and induced flow separation delay were observed in the supercritical Re regime. Flow separation occurred on the sides of the body later than on the top, regardless of whether Re was subcritical or supercritical. The spatial and temporal features of flow structures at different scales were described in detail based on the present high-resolution simulations. The coexistence of lateral oscillations and arch-type vortex shedding occurred throughout the subcritical and supercritical Re range. However, both of these motions diminished in scale and strength at supercritical Re. Flow motion frequencies were also quantified. The frequency ratio of arch vortex shedding to lateral oscillation was approximately 4 at subcritical Re but decreased to 3 at supercritical Re.
The wind pressure coefficient (Cp) can be divided into a mean wind pressure coefficient (Cp, mean) and peak wind pressure coefficient (Cp, peak). In this study, the Cp, mean and Cp, peak values of single-span greenhouses were estimated using time-dependently computed large eddy simulation (LES) of the computational fluid dynamics (CFD) model. First, the Cp, mean and Cp, peak values of single-span greenhouses were measured through a wind tunnel test considering various experimental conditions, such as the roof slope and roof curvature radius. The CFD-computed Cp, mean and Cp, peak values were compared with those measured in the wind tunnel for the validation of the CFD model. Especially, the computed Cp, mean values using the LES model were compared with the computed Cp, mean values using the RANS model performed by Kim, Lee, and Kwon (2017). Consequently, the LES model predicted the Cp, mean values more accurately than those predicted by the RANS. In addition, the Cp, peak values computed using the LES model showed only fine differences with those measured in some other regions; however, the overall tendency was similar between the measured and computed Cp, peak values. From these results, it was determined that the LES model can properly predict the Cp, peak values as well as the Cp, mean values.
The computational fluid dynamics (CFD) method is widely used today for determining the wind action on structures. Along with the CFD method the wind tunnel experiments are also employed to achieve this purpose. The cost and effort required for a wind tunnel test, however, is much higher than that of the CFD method; therefore, it is beneficial to extend and validate the use of CFD method to various classes and types of structures. This paper firstly compares the results of CFD method with the results of three series of wind tunnel tests available in the literature. Then, numerically the effect of structural flexibility and the neighbourhood of the objects on the wind pressure distribution coefficients are studied. Two and three half spheres are arranged sequentially and transversally. According to the CFD analysis, at a clear distance greater than 2.5 times the diameter of the (obstructing) dome, the wind dynamic interference effects on the reference hemisphere are diminished.
A common type of failure observed during hurricane events is the detachment of the roofs of light-frame wood houses because of the inability of the roof-to-wall connection (RTWC) to transfer uplift wind loads acting on the roofs. This paper introduces a computationally efficient procedure for conducting nonlinear analysis of the roofs to determine the connections forces using a semi-analytical solution. The new procedure simulates an entire roof as a continuous beam resting on elastic supports through the use of statically indeterminate slope deflection equations that include shear deformations. It accounts for the two-dimensional spatial variation of the uplift wind pressure. The new model was validated against a three-dimensional finite element model and also through a comparison with the results of experimental testing of a gable roof house previously conducted at the University of Western Ontario, Canada. The model was used to determine the load sharing among the trusses under a realistic load. It was also used to assess the adequacy of using the tributary area method under a code load. Based on the validation results, the semi-analytical solution model demonstrated an ability to predict the response of a roof under uplift wind loads that vary both in time and space. The main advantage of the new model is the efficient computational time compared to three-dimensional finite element modeling. As such this model can be used in the future for conducting probabilistic analysis of the roof taking into account the variation of the connection properties.
Because aboveground storage tanks are usually empty during construction, inspection and repair periods, they are more vulnerable to buckling due to wind loading than they are in use filled with liquid product. The uplift effect of the bottom plate of empty storage tanks due to wind pressure is investigated in this work. The buckling behavior of two sets of tanks were studied using finite element analysis (FEA): (1) tanks with bottom plate modeled directly in the analysis and (2) tanks without bottom plate but suitable boundary conditions applied in the analysis. Linear bifurcation analysis (LBA) and geometrically nonlinear analysis including imperfections (GNIA) are performed to obtain the deformation shapes and buckling capacities of tanks. The soil supporting the with-bottom tanks were modeled using nonlinear compression only springs connected to each node of the bottom plate. The buckling modes and the results from GNIA show that when a bottom is included into a tank FEA simulation, instead of fixed, the buckling pressure load will be substantially reduced for slender tanks having height to diameter ratio greater than 1/3. However, the broad tanks having height to diameter ratio less than 1/3 experience relatively small reduction in buckling load.
This study investigates the characteristics of wind pressures on retractable dome roofs. Wind tunnel tests were conducted on spherical dome roofs with various wall height-span ratios (0.1–0.5) and opening ratios (0%, 10%, 30% and 50%). After the characteristics of mean and peak pressure coefficients were examined, the peak pressure coefficients on closed dome roofs were compared with those of the current Japanese wind load code (AIJ-RLB (2015)), and peak pressure coefficients on dome roofs with openings for cladding design were proposed. The analysis results can be summarized as follows. As wall height-span ratio increases, the absolute values of mean and minimum pressure coefficients increase, and the absolute values of minimum pressure coefficients on the outermost interior of open dome roofs increase with opening ratio. For cladding design of dome roofs with openings, dome roofs were divided into two zones, and negative and positive peak pressure coefficients were proposed based on experimental results.
Understanding the characteristics of internal and external pressures on structures is important for structural design. This paper has experimentally observed internal and external pressures on long-span domes and investigated their non-Gaussian characteristics. Wind tunnel tests on internal and external pressure measurements for a single dome and two neighboring domes were performed. The distribution, non-Gaussian characteristics, and non-Gaussian peak factor of the internal and external wind pressure on the single dome were analyzed. After the results were validated, the distribution, non-Gaussian characteristics, and non-Gaussian peak factor of the internal and external wind pressure on the two neighboring domes under different wind attack angles (interference effect) were discussed. The results show that the characteristics of internal pressures on the single dome and the two neighboring domes are different from that of external pressures, and both the internal and external pressure show remarkable non-Gaussian characteristics. According to these characteristics, positive and negative areas of the internal and external pressure on the domes were defined, and the non-Gaussian peak factors were determined. This study has not only advanced our understanding on characteristics of internal and external pressure on long-span domes, but determined non-Gaussian peak factors that are used for structural design.
This paper numerically studies the wind effect on grooved and scallop domes. The introduction of a groove on a spherical dome causes abrupt change on its wind pressure coefficient (Cp) in the vicinity of the groove. The sharpness of indentation varies with the position angle of the axis of the groove to wind direction and obtains its highest effect at 90°. This paper develops equations for the distribution of Cp on the surface of spherical, grooved and scallop domes with rise to span ratios [0, 0.7]. The results of the equations agree reasonably well with that of CFD.
Steel cylindrical tanks are formed by very thin-walled shells and as typical thin-walled structures, they are very sensitive to buckling under external pressures, especially when they are empty or at low liquid level. Results of numerical investigations on the effects of spiral stairway on the buckling behavior of ground based steel cylindrical liquid storage tanks subjected to both wind and vacuum pressures are presented. It is concluded that by choosing appropriate circumferential position for stairway, considering regional prevailing wind direction, designers can use the noticeable amount of added strength arising from stairway as a safety margin. In other words, it seems that the spiral stairway acts as an oblique stiffener on the tank wall. Contrary to the case of wind loading, the stairway has negligible effect on buckling resistance of tanks under vacuum pressure. However, it changes the buckling modes of tank significantly under vacuum loading.
Wind-induced pressures are measured on a group of six cooling towers during a series of boundary layer wind tunnel tests, which varied among two tower arrangements, three tower spacings and sixteen wind directions. Based on the statistical results of wind pressures on the throat level, twenty-seven types of pressure distribution curves including nine symmetrical distributions and eighteen asymmetrical distributions are proposed and fitted by the fifteen-term trigonometric function with seven sine terms, seven cosine terms and one constant term. The local buckling factor and construction cost are adopted to compare wind effects induced by the twenty-seven types of wind loads. The results show that there are three symmetrically distributed and three asymmetrically distributed wind loads, which can lead to more severe wind effects. The most adverse wind load pattern of all is the symmetrical distribution with the largest pressures around the stagnation point area, largest suctions in both sideward regions and lowest suctions in the leeward region.
This paper presents the wind force coefficients on non-conventional building of the free form shell structure generated by the simulation of a membrane initially in the horizontal plane surface with hexagonal plant, under the action of self-weight and supported on six corners. Initially is presented the computational model for free form shells generation, the model generated and the correspondent construction of the hexagonal free form shell scale model in fiberglass by a CAD/CAM system. Following is presented the wind tunnel of atmospheric boundary layer and the scale model used to perform the tests to obtain the wind force coefficients. Finally are presented the numerical analysis of the scale model of free form shell structure by the computational fluid dynamics software ANSYS-CFX. The wind force, coefficients for the free form shell obtained by the wind tunnel tests and ANSYS-CFX software are compared. The results demonstrates the possibilities to apply the computational fluid dynamic analysis for future applications to analyze other free form shells in order to obtain wind force coefficients.
The full-scale measurements are compared with the wind tunnel test results for the long-span roof latticed spatial structure of Shenzhen Citizen Center. A direct comparison of model testing results to full-scale measurements is always desirable, not only in validating the experimental data and methods but also in providing better understanding of the physics such as Reynolds numbers and scale effects. Since the quantity and location of full-scale measurements points are different from those of the wind tunnel tests taps, the weighted proper orthogonal decomposition technique is applied to the wind pressure data obtained from the wind tunnel tests to generate a time history of wind load vector, then loads acted on all the internal nodes are obtained by interpolation technique. The nodal mean wind pressure coefficients, root-mean-square of wind pressure coefficients and wind pressure power spectrum are also calculated. The time and frequency domain characteristics of full-scale measurements wind load are analyzed based on filtered data-acquisitions. In the analysis, special attention is paid to the distributions of the mean wind pressure coefficients of center part of Shenzhen Citizen Center long-span roof spatial latticed structure. Furthermore, a brief discussion about difference between the wind pressure power spectrum from the wind tunnel experiments and that from the full-scale in-site measurements is compared. The result is important fundament of wind-induced dynamic response of long-span spatial latticed structures.
In the past 50 years, Computational Wind Engineering (CWE) has undergone a successful transition from an emerging field into an increasingly established field in wind engineering research, practice and education. This paper provides a perspective on the past, present and future of CWE. It addresses three key illustrations of the success of CWE: (1) the establishment of CWE as an individual research and application area in wind engineering with its own successful conference series under the umbrella of the International Association of Wind Engineering (IAWE); (2) the increasing range of topics covered in CWE; and (3) the history of overview and review papers in CWE. The paper also outlines some of the earliest achievements in CWE and the resulting development of best practice guidelines. It provides some views on the complementary relationship between reduced-scale wind-tunnel testing and CFD. It re-iterates some important quotes made by CWE and/or CFD researchers in the past, many of which are still equally valid today and which are provided without additional comments, to let the quotes speak for themselves. Next, as application examples to the foregoing sections, the paper provides a more detailed view on CFD simulation of pedestrian-level wind conditions around buildings, CFD simulation of natural ventilation of buildings and CFD simulation of wind-driven rain on building facades. Finally, a nonexhaustive perspective on the future of CWE is provided.
The greenhouse type metal structures are increasingly used in modern construction of livestock farms because they are less laborious to construct and they provide a more favorable microclimate for the growth of animals compared to conventional livestock structures. A key stress factor for metal structures is the wind. The external pressure coefficient () is used for the calculation of the wind effect on the structures. A high pressure coefficient value leads to an increase of the construction weight and subsequently to an increase in the construction cost. The EC1 in conjunction with EN 13031-1:2001, which is specialized for greenhouses, gives values for this coefficient. This value must satisfy two requirements: the safety of the structure and a reduced construction cost. In this paper, the Navier - Stokes and continuity equations are solved numerically with the finite element method (Galerkin Method) in order to simulate the two dimensional, incompressible, viscous air flow over the vaulted roofs of single span and twin-span with eaves livestock greenhouses' structures, with a height of 4.5 meters and with length of span of 9.6 and 14 m. The simulation was carried out in a wind tunnel. The numerical results of pressure coefficients, as well as, the distribution of them are presented and compared with data from Eurocodes for wind actions (EC1, EN 13031-1:2001). The results of the numerical experiment were close to the values given by the Eurocodes mainly on the leeward area of the roof while on the windward area a further segmentation is suggested.
No parametric data exists in the main international standards to calculate wind action on buildings with hyperbolic paraboloid roofs. This type of roof, however, is particularly efficient in covering medium to large spans and is highly competitive compared to traditional structures. Our research aims at parameterization of pressure coefficients on roofs with hyperbolic paraboloid shapes with four different footprints, respectively, square, rectangular, circular and elliptical.This paper studies the elliptical shape in particular. Aerodynamic wind tunnel tests have been performed on these shapes with the object of calculating pressure coefficients. A comparison is also made between elliptical and circular shapes in order to demonstrate the high efficiency of the elliptical shape.
A series of wind tunnel tests was performed to investigate the effects of Reynolds number on the aerodynamic characteristics of hemispherical dome in smooth and turbulent boundary layer flows. Reynolds number of this study varies from 5.3×104 to 2.0×106. Instantaneous pressures were measured through high frequency electronic scanner system. Mean and RMS pressure coefficients on the center meridian and the overall pressure patterns of domes were calculated for comparative study. The results indicate that, in smooth flow, the transition of separation flow occurs in-between Re=1.8×105–3.0×105 and pressure distributions become relatively stable after Re>3.0×105. In turbulent flow, the transition of separation flow occurs at lower Reynolds number, Re
Milad tower is located in Tehran, Iran, and is a 436-m telecommunication tower ranked as the fourth tallest structure in the world. Because of its specific use and also because of highly sensitive communication devices installed on the tower, nonlinear deformations under future severe winds and earthquakes should be studied. In this paper, a comprehensive study is carried out to investigate the effect of wind on this tower. The techniques of computational fluid dynamics, such as large eddy simulation (LES), Reynolds Averaged Navier–Stokes Equations (RANS) model and so on, were adopted in this study to predict wind loads on and wind flows around the building. The calculated results are compared with those of wind tunnel test. It was found through the comparison that the LES with a dynamic subgrid-scale model can give satisfactory predictions for mean and dynamic wind loads on the specific structure of Milad tower, while the RANS model with modifications can yield encouraging results in most cases and has the advantage of providing rapid solutions. Furthermore, it was observed that typical features of the flow fields around such a surface-mounted bluff body standing in atmospheric boundary layers can be captured numerically. Copyright © 2009 John Wiley & Sons, Ltd.