• Wind and Structures
• /
• 제26권1호
• /
• pp.11-24
• /
• 2018

Full-scale measurements of wind action on the open roof structure of the WuXi grand theater, which is composed of eight large-span free-form leaf-shaped space trusses with the largest span of 76.79 m, were conducted during the passage of Typhoons HaiKui and SuLi. The wind pressure field data were continuously and simultaneously monitored using a wind pressure monitoring system installed on the roof structure during the typhoons. A detailed analysis of the field data was performed to investigate the characteristics of the fluctuating wind pressure on the open roof, such as the wind pressure spectrum, spatial correlation coefficients, peak wind pressures and non-Gaussian wind pressure characteristics, under typhoon conditions. Three classical methods were used to calculate the peak factors of the wind pressure on the open roof, and the suggested design method and peak factors were given. The non-Gaussianity of the wind pressure was discussed in terms of the third and fourth statistical moments of the measured wind pressure, and the corresponding indication of the non-Gaussianity on the open roof was proposed. The result shows that there were large pulses in the time-histories of the measured wind pressure on Roof A2 in the field. The spatial correlation of the wind pressures on roof A2 between the upper surface and lower surface is very weak. When the skewness is larger than 0.3 and the kurtosis is larger than 3.7, the wind pressure time series on roof A2 can be taken as a non-Gaussian distribution, and the other series can be taken as a Gaussian distribution.

• Wind and Structures
• /
• 제4권1호
• /
• pp.73-84
• /
• 2001

Analysis of pressures measured on the roof of the full-scale Texas Tech building and a 1/50 scale model of a typical house showed that the pressure fluctuations on cladding fastener and cladding-truss connection tributary areas have similar characteristics. The probability density functions of pressure fluctuations on these areas are negatively skewed from Gaussian, with pressure peak factors less than -5.5. The fluctuating pressure energy is mostly contained at full-scale frequencies of up to about 0.6 Hz. Pressure coefficients, $C_p$ and local pressure factors, $K_l$ given in the Australian wind load standard AS1170.2 are generally satisfactory, except for some small cladding fastener tributary areas near the edges.

• Wind and Structures
• /
• 제4권6호
• /
• pp.481-494
• /
• 2001

Wind tunnel pressure tests were conducted on a 1:100 scale model of a large industrial building with solar panels mounted parallel to the flat roof. The model form was chosen to have the same aspect ratio as the Texas Tech University test building. Pressures were simultaneously measured on the roof, and on the topside and underside of the solar panel, the latter two combining to produce a nett panel pressure. For the configurations tested, varying both the lateral spacing between the panels and the height of the panels above the roof surface had little influence on the measured pressures, except at the leading edge. The orientation of the panels with respect to the wind flow and the proximity of the panels to the leading edge had a greater effect on the measured pressure distributions. The pressure coefficients are compared against the results for the roof with no panels attached. The model results with no panels attached agreed well with full-scale results from the Texas Tech test building.

• Wind and Structures
• /
• 제36권6호
• /
• pp.393-404
• /
• 2023

Skewness and kurtosis are important higher-order statistics for simulating non-Gaussian wind pressure series on low-rise buildings, but their predictions are less studied in comparison with those of the low order statistics as mean and rms. The distribution gradients of skewness and kurtosis on roofs are evidently higher than those of mean and rms, which increases their prediction difficulty. The conventional artificial neural networks (ANNs) used for predicting mean and rms show unsatisfactory accuracy in predicting skewness and kurtosis owing to the limited capacity of shallow learning of ANNs. In this work, the deep neural networks (DNNs) model with the ability of deep learning is introduced to predict the skewness and kurtosis on a low-rise building. For obtaining the optimal generalization of the DNNs model, the hyper parameters are automatically determined by Bayesian Optimization (BO). Moreover, for providing a benchmark for future studies on predicting higher order statistics, the data sets for training and testing the DNNs model are extracted from the internationally open NIST-UWO database, and the prediction errors of all taps are comprehensively quantified by various error metrices. The results show that the prediction accuracy in this study is apparently better than that in the literature, since the correlation coefficient between the predicted and experimental results is 0.99 and 0.75 in this paper and the literature respectively. In the untrained cornering wind direction, the distributions of skewness and kurtosis are well captured by DNNs on the whole building including the roof corner with strong non-normality, and the correlation coefficients between the predicted and experimental results are 0.99 and 0.95 for skewness and kurtosis respectively.

• 생물환경조절학회지
• /
• 제2권1호
• /
• pp.46-52
• /
• 1993

The wind pressure distributions were analyzed through the wind tunnel experiment to provide fundamental criteria for the structural design on the three-span arched house according to the wind directions. In order to investigate the wind force distribution, the variation of the wind force coefficients, the mean wind force coefficients, the drag force coefficients and the lift force coefficients were estimated from the experimental data. The results obtained are as follows : 1. The variation of the wind force with the wind directions on the side walls was the greatest at the upwind edge of the walls. The change of pressure from the positive to the negative on the side walls occurred at the wind direction of 30$^{\circ}$ in the first house and 60$^{\circ}$ in the third house. 2. The maximum negative wind force along the length of the roof appeared at the length ratio of 0-0.2, when the wind directions were 90$^{\circ}$ in the first house, 60$^{\circ}$ in the second house and 30$^{\circ}$ in the third house. 3. The maximum negative wind force along the width of the roof appeared at the width ratio and the wind direction of 0.4 and 0$^{\circ}$ in the first house, 0.4-0.6 and 30$^{\circ}$ in the second house and 0.6 and 30$^{\circ}$ in the third house, respectively. 4. The maximum mean positive and negative wind forces occurred at the wind direction of 60$^{\circ}$ and 30$^{\circ}$, respectively, on the side walls of the first house, and the maximum mean negative wind force on the roof occurred at the wind direction of 30$^{\circ}$ in third house. 5. The maximum drag and lift forces occurred at the wind direction of 30$^{\circ}$, and the maximum lift force appeared in the third house. 6. The parts to be considered for the local wind forces were the edges of the walls, the edges of the x-direction of the roofs, and the locations of the width ratio of 0.4 of the first and third house and the center of the width of the second house for the y-direction of the roofs.

• 생물환경조절학회지
• /
• 제4권1호
• /
• pp.1-8
• /
• 1995

One of the most destructive forces around greenhouses is wind. Wind loads can be obtained by multiplying velocity pressure by dimensionless wind force coefficient. Generally, wind force coefficients can be determined by wind tunnel experiments. The wind force coefficient distribution on a single - span arched greenhouse was estimated using experimental data and compared with reported values from various countries. The results obtained are as follows : 1. The coefficients obtained from this study agree with the values proposed by G. L. Nelson except about 0.5 of difference in the middle region of roof section. This discrepancy is mainly attributed to the dissimilarity of experimental conditions (or wind tunnel test such as Reynolds number, type of terrain, surface roughness of model, location of the lapping and measuring methods. 2. Considering that the wind force coefficients are varied along the height of a wall at wind direction perpendicular to wall, structural analysis using subdivided wind force coefficient distribution is more resonable for wall. 3. It is recommendable that wind force coefficient distribution on a roof should take more subdivision than the existing four equal divisions for more accurate structural design. 4. Structural design using wind forces close to real values is more advantageous in safety and expense.

• Wind and Structures
• /
• 제38권3호
• /
• pp.203-214
• /
• 2024

Large-span cantilevered roof represents a unique type of structure that is vulnerable to wind loads. Inspired by the need to maximumly reducing the rooftop wind loads, this study examined the feasibility of positioning vented slots on the leading edge, and the effectiveness of such aerodynamic mitigation measures are assessed via both physical and numerical simulations. The reliability of numerical simulation was evaluated via comparisons with the wind tunnel tests. The results indicated that, the variation of venting hole arrangement can cause significant change in the rooftop wind load characteristics. For the cases involved in this study, the maximum reduction of mean and peak wind suction coefficients are found to be 9% and 8% as compared to the original circular slot without venting holes. In addition, the effect of slot shape is also evident. It was shown that the triangular shaped slot tends to increase the wind suction near the leading edge, whereas the hexagonal and octagonal shaped slots are found to decrease the wind suction. In particular, with the installation of octagonal shaped slot, the maximum reduction of wind suction coefficients near the leading edge reaches up to 31% as compared to the circular shaped slot, while the maximum reduction of mean wind suction coefficients is about 30%.

• Wind and Structures
• /
• 제18권2호
• /
• pp.155-180
• /
• 2014

The safety of road vehicles on the ground in crosswind has been investigated for many years. One of the most important fundamentals in the safety analysis is aerodynamic characteristics of a vehicle in crosswind. The most common way to study the aerodynamic characteristics of a vehicle in crosswind is wind tunnel tests to measure the aerodynamic coefficients and/or pressure coefficients of the vehicle. Due to the complexity of wind tunnel test equipment and procedure, the features of flow field around the vehicle are seldom explored in a wind tunnel, particularly for the vehicle moving on the ground. As a complementary to wind tunnel tests, the numerical method using computational fluid dynamics (CFD) can be employed as an effective tool to explore the aerodynamic characteristics of as well as flow features around the vehicle. This study explores crosswind effects on a high-sided lorry on the ground with and without movement through CFD simulations together with wind tunnel tests. Firstly, the aerodynamic forces on a stationary lorry model are measured in a wind tunnel, and the results are compared with the previous measurement results. The CFD with unsteady RANS method is then employed to simulate wind flow around and wind pressures on the stationary lorry. The numerical aerodynamic forces are compared with the wind tunnel test results. Furthermore, the same CFD method is extended to investigate the moving vehicle on the ground in crosswind. The results show that the CFD results match with wind tunnel test results and the current way using aerodynamic coefficients from a stationary vehicle in crosswind is acceptable. The CFD simulation can provide more insights on flow field and pressure distribution which are difficult to be obtained by wind tunnel tests.

• Wind and Structures
• /
• 제24권2호
• /
• pp.95-118
• /
• 2017

The present study revealed comparison the pressure distribution on the surfaces of regular cross plan shaped building with angular cross plan shaped building which is being transformed from basic cross plan shaped building through the variation of internal angles between limbs by $15^{\circ}$ for various wind incidence angle from $0^{\circ}$ to $180^{\circ}$ at an interval of $30^{\circ}$. In order to maintain the area same the limbs sizes are slightly increased accordingly. Numerical analysis has been carried out to generate similar nature of flow condition as per IS: 875 (Part -III):1987 (a mean wind velocity of 10 m/s) by using computational fluid dynamics (CFD) with help of ANSYS CFX ($k-{\varepsilon}$ model). The variation of mean pressure coefficients, pressure distribution over the surface, flow pattern and force coefficient are evaluated for each cases and represented graphically to understand extent of nonconformities due to such angular modifications in plan. Finally regular cross shaped building results are compared with wind tunnel results obtained from similar '+' shaped building study with similar flow condition. Reduction in along wind force coefficients for angular crossed shaped building, observed for various skew angles leads to develop lesser along wind force on building compared to regular crossed shaped building and square plan shaped building. Interference effect within the internal faces are observed in particular faces of building for both cases, considerably. Significant deviation is noticed in wind induced responses for angular cross building compared to regular cross shaped building for different direction wind flow.

• Wind and Structures
• /
• 제28권2호
• /
• pp.89-97
• /
• 2019

The non-linear equations governing wind-induced internal pressures for a two-compartment building with background leakage are linearized based on some reasonable assumptions. The explicit admittance functions for both building compartments are derived, and the equivalent damping coefficients of the coupling internal pressure system are iteratively obtained. The RMS values of the internal pressure coefficients calculated from the non-linear equations and linearized equations are compared. Results indicate that the linearized equations generally have good calculation precision when the porosity ratio is less than 20%. Parameters are analyzed on the explicit admittance functions. Results show that the peaks of the internal pressure in the compartment without an external opening (Compartment 2) are higher than that in the compartment with an external opening (Compartment 1) at lower Helmholtz frequency. By contrast, the resonance peak of the internal pressure in compartment 2 is lower than that in compartment 1 at higher Helmholtz frequencies.