Wind Effects on Non-Standard Shapes and Structures

Dataset Description

The data contained in this package aims to act as a open-access database with pressure maps, alongside proper description of the situations considered. In addition, the data within this package aims to contribute to the development and validation of an open-source numerical project with predictive capabilities. The study replicates various complex shapes of membrane structures for testing under Atmospheric Boundary Layer (ABL) flow, Tornadic-like flow, and Downburst-like flow at the Wind Engineering, Energy, and Environment (WindEEE) research facility. Prestressed membrane structures are lightweight structures with a multitude of application areas from small shading devices and façade elements to wide-span roofs covering sport stadia and other leisure facilities. Due to their architectural appeal and the ability to cover large areas with minimal material usage, they represent a highly resource-efficient category of engineering structures.

S1. Wind Field Characterization

The purpose of this specimen was to develop and identify an ABL profile that corresponds to the 1:25 model scale. Tests were performed with various turbulence levels created using a combination of roughness elements, spires, and a trip. For each test, 3-dimensional point-based velocity measurements were taken with TFI Cobra Probes covering a range of heights.

E1. Atmospheric Boundary Layer (ABL) development

Given the model scale, size, and region of installation, ABL flows were tested to develop an accurate simulation for testing the membrane structure models under. Tests were performed with both low-speed and high-speed configurations. Each configuration involved measurements with various turbulence production methods such as roughness elements, spires, trip, and fan speed variation.

S2. M1 (Hyperbolic Paraboloid) Pressure Model

The purpose of this specimen was to test the M1, doubly, curved geometry at a scale of 1:25 under ABL, Downburst, and Tornado winds. The full-scale size of the geometry is 6 x 6 x 5m and at a scale of 1:25 the tested model is 24 x 24 x 20cm where the canopy itself is 8cm tall beginning 12cm from the ground. The canopy is 1cm thick overall (offset 0.5cm above and below the original surface geometry) with a 0.6cm cavity in between for pressure tubing.

E1. ABL Loading

This test involved the M1 model described above subjected to low-speed ABL, and high-speed ABL. The model was tested every 10° between 0 and 180° plus a 45° angle of attack.

E2. Downburst Loading

This test involved the M1 model tested under downburst-like flow loading conditions in three configurations where the model is offset from the center of the turntable to achieve 3 r/D values: 0.8, 1.0, and 1.2 (in the -x direction) (offset 0.65m, 1.29m and 1.95m respectively).

E3. Tornado Loading

This test involved the M1 model tested under tornado-like flow loading conditions in two configurations where the model is offset from the center of the turntable by 0cm and 25cm (in the -x direction).

S3. M2 (RidgeValley) Pressure Model

The purpose of this specimen was to test the M2 (RidgeValley) doubly curved geometry at a scale of 1:25 under ABL winds. The equilibrium shape of the M2 (RidgeValley) can be determined via form finding for a pre-stress ratio of 4 (e.g. kN/m) as an isotropic membrane pre-stress to 30 (e.g. kN) in the edge and ridge cables, with fixed support points. The full-scale size of the geometry is 6 x 6 x 5m and at a scale of 1:25 the tested model is 24 x 24 x 20cm where the canopy itself is 8cm tall beginning 12cm from the ground. The canopy is 1cm thick overall (offset 0.5cm above and below the original surface geometry) with a 0.6cm cavity in between for pressure tubing.

E1. ABL Loading

This test involved the model described above subjected to low-speed ABL, and high-speed ABL. The model was tested every 10° between 0 and 180° plus a 45° angle of attack.

S4. M3 (Arch-Supported) Pressure Model

The purpose of this specimen was to test the M3 (Arch-Supported) doubly curved geometry at a scale of 1:25 under ABL winds. The equilibrium shape of the M3 can be determined via form finding for a pre-stress ratio of 4 (e.g. kN/m) as an isotropic membrane pre-stress to 30 (e.g. kN) in the edge cables, with fixed line supports at the arches. The arch geometry is defined as a NURBS curve (p=3, CPs at (0,0,0), (0,2,2.65), (0,4,2.65), (0,6,0), knot vector (0,0,0,1,1,1)). The full-scale size of the geometry is 6 x 6 x 5m and at a scale of 1:25 the tested model is 24 x 24 x 20cm where the canopy itself is 8cm tall beginning 12cm from the ground. The canopy is 1cm thick overall (offset 0.5cm above and below the original surface geometry) with a 0.6cm cavity in between for pressure tubing.

E1. ABL Loading

This test involved the model described above subjected to low-speed ABL, and high-speed ABL. The model was tested every 10° between 0 and 180° plus a 45° angle of attack.

S5. M4 (Cone) Pressure Model

The purpose of this specimen was to test the M4 (Cone) doubly curved geometry under ABL winds in an isolated, 1x3 row array, and a 3x3 square array configuration. The equilibrium shape of the M4 can be determined via form finding for an isotropic membrane pre-stress, with fixed supports at top and bottom circles. The support circle radii are 0.8m and 4.24m (full-scale). The form found shapes are then intersected at a 6m distance in order to generate straight edges for closed array geometries. The full-scale size of the isolated geometry is 6 x 6 x 5m and at a scale of 1:25 the tested model is 24 x 24 x 20cm where the canopy itself is 8cm tall beginning 12cm from the ground. The canopy is 1cm thick overall (offset 0.5cm above and below the original surface geometry) with a 0.6cm cavity in between for pressure tubing.

E1. ABL Loading Stand Alone

This test involved the model described above subjected to low-speed ABL, and high-speed ABL. The model was tested every 10° between 0 and 180° plus a 45° angle of attack.

E2. ABL Loading Row (1 x 3) Array

This test involved the M3 model plus 8 similarly shaped, adjacent dummy models arranged in 3 different configurations subjected to low-speed ABL, and high-speed ABL. The configurations include of a 1x3 line relative to the main pressure model. In the configurations, the pressure model is located: at the center of the line, and at the corner of the line. The model was tested every 10° between 0 and 180° plus a 45° angle of attack.

E3. ABL Loading Square (3 x 3) Array

This test involved the M3 model plus 2 similarly shaped, adjacent dummy models arranged in 2 different configurations subjected to low-speed ABL, and high-speed ABL. The configurations include different locations of the main pressure model within a 3x3 square array. In the configurations, the pressure model is located: at the center of the grid, at the corner of the grid, and at the center edge of the grid. The model was tested every 10° between 0 and 180° plus a 45° angle of attack.

S6. M5 (Umbrella) Pressure Model

The purpose of this specimen was to test the M5 (Umbrella) doubly curved geometry at a scale of 1:25 under ABL winds. The equilibrium shape of the M5 can be determined via formfinding for a pre-stress ratio of 4 (e.g. kN/m) as an isotropic membrane pre-stress to 30 (e.g. kN) in the edge cables, with fixed line supports at the bottom circle (radius of 0.8m in full scale) and point supports at the top corners. The full-scale size of the geometry is 6 x 6 x 5m and at a scale of 1:25 the tested model is 24 x 24 x 20cm where the canopy itself is 8cm tall beginning 12cm from the ground. The canopy is 1cm thick overall (offset 0.5cm above and below the original surface geometry) with a 0.6cm cavity in between for pressure tubing.

E1. ABL Loading

This test involved the model described above subjected to low-speed ABL, and high-speed ABL. The model was tested every 10° between 0 and 180° plus a 45° angle of attack.