RSA 2015 - Wind - Simulator
RSA 2015 - Wind - Simulator
RSA 2015 - Wind - Simulator
TM
Autodesk Robot Structural Analysis Professional 2015 software incorporates a new wind simulation tool that enables
users to test their designs in a virtual wind tunnel. Utilizing an intuitive user interface, designers can quickly and easily
apply wind flows to a structure and get nearly instant results to either visualize surface pressures or generate wind
loads to be used in further design and analysis. To accomplish this, the tool incorporates computational fluid dynamics
(CFD) into a streamlined workflow practical for design-phase analysis. The following document includes details and
results of a validation study comparing results from the softwares wind simulation to physical results obtained from
wind tunnel testing.
www.autodesk.com/robot
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Contents
Introduction ..................................................................................................................................................................3
The designers wind simulator ........................................................................................................................................3
Virtual wind tunnel validation ......................................................................................................................................4
Wind tunnel test..............................................................................................................................................................5
Simulation.......................................................................................................................................................................6
Specific validation objectives ..........................................................................................................................................7
Validation results .........................................................................................................................................................8
How accurately are flow effects simulated?....................................................................................................................8
What is the effect of simulation duration? .....................................................................................................................11
Conclusions ................................................................................................................................................................14
Appendix A Pressure tap locations .......................................................................................................................15
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Introduction
In the course of a building project, designers must consider
the impact of wind. Whether simply determining design loads
or trying to optimize the aerodynamic effects of the building
structure and surroundings, wind analysis can pose a
challenge.
Although detailed codes and standards exist for determining
minimum requirements, their provisions are limited to
generalized conditions. Projects that venture from the basic
parameters of these standards often require wind tunnel
testing. In fact, many codes recognize the need for testing to
provide more accurate design evaluation in areas where
generalized standards may be inadequate.
In fact, wind tunnel testing is a key step in the process of the design and analysis of a building, whether it is
required due to height or geometric and/or environmental complexity or simply to optimize the cost and safety of
the project. However, testing is expensive and interpreting results can be time-consuming. Due to these
challenges, it can be impractical to use a testing tool in the design development phase of a project; yet, that is
where an understanding of wind effects can have significant benefit and impact to the projects design solution.
So what about simulation? Computational fluid dynamics can be used to provide very detailed analysis of the fluid
flow induced by wind. However, CFD analyses are characteristically complex and typically require software
outside of the scope of the design engineers expertise or workflow.
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Complex geometryA high-rise building (HRB) of atypical geometry (not standard code shapes) was
chosen. The model incorporated strategic features, including multiple towers of staggered height, gable
roofs, and a pedestal building base.
Simple workflowThe wind tunnel test was conducted using standard processes and the simulation was
performed using standard workflows available to users of Robot Structural Analysis 2015.
The goal of the study was to evaluate the level of accuracy that could be expected during a typical application. The
physical and simulated tests are summarized below and described in greater detail in the sections that follow.
Experiment
Simulation
Model
Wind Tunnel
3 m wide x 14 m long
Dimensions
Wind Profile
Same
Same
Same**
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Wind profile
Testing was conducted using wind velocity and turbulence intensity profiles from two different terrain categories per
ISO4254:2008 (wind actions on structures)TC 1.5 (relatively flatdeserts and plains) and TC 3.0 (suburban and
forest terrain)and were based on a 20 m/s hourly mean wind speed at a building height of 81 m.
Wind direction was achieved by rotating the model. Tests were conducted every 15 degrees, resulting in 24 tests for
each terrain category.
Data collection
The model was instrumented with 511 pressure
sensors, including 114 on the roof surfaces. A
view of the pressure tap location on the north
elevation and roof are shown here for example.
Larger diagrams for each face can be found in
Appendix A.
The measurement sample time corresponded to
a full-scale sample time of approximately 68
minutes.
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Simulation
For purposes of comparison, the wind tunnel test was duplicated in the simulation to the most practical degree
possible while keeping within the typical workflow of the Robot Structural Analysis user.
Study model
A CAD model (STL) of the full-scale building was used for the simulation. The building dimensions are 84 m wide and
81 m high.
Figure 4. Autodesk Robot Structural Analysis with building model and wind simulation input window
open.
Wind profile
The software user interface enables the user to input a velocity profile to simulate a boundary layer wind flow. As
such, velocity profiles were entered and applied to correlate with the TC 1.5 and TC 3.0 terrain categories that were
selected for the wind tunnel test and were similarly based on a 20 m/s hourly mean at a reference building height of
81 m.
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Figure 5. The terrain category 1.5 wind profile was translated into several points used to input a velocity profile into the Robot
Structural Analysis wind simulation tool.
As with the physical test, wind direction was incremented every 15 degrees for a total of 24 simulations for each
terrain category.
Data collection
In practice, the wind tunnel simulator in Robot Structural Analysis provides the user with a visualization of surface
pressure contours and produces resultant wind loads for purposes of structural analysis. For direct comparison
against the physical wind tunnel data, pressure data was extracted at specific coordinates to match the placement of
the pressure taps on the physical model.
Simulations were run for approximately 70 seconds (simulated time) to evaluate effects of the transient solver on
results.
How accurately are flow effects simulated? Measurements were examined for an example case to
evaluate the accuracy of the simulation in predicting flow patterns and pressure levels.
What is the effect of simulation duration? The wind tunnel simulator is a transient solver. Results were
examined to determine accuracy of results over time and the efficacy or influence of setting the deviation
parameter.
o
How does the error vary over time? What appears to be the optimal time to run a simulation?
The results of these studies would give users an idea of the level of accuracy to be expected during a typical
application and an understanding of how the deviation parameter affects results.
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Validation results
How accurately are flow effects simulated?
To begin, we examine in general how a single case performs. Wind at the east face was selected to investigate wind
flow around back to back towers. In this case, we consider a simulation in which the simulation is run until the default
load deviation factor of 0.5% is reachedin this case, 12.89 seconds.
Tower 2
Tower 1
Figure 6. Results for 20 m/s wind at building east face for Robot Structural Analysis wind simulator (left) and Windtech wind tunnel
(right).
The above figures show a comparison of mean surface pressure contours for the Robot Structural Analysis simulation
(left) and wind tunnel results (right) for wind from the east direction. (Note that the color assignments of the contour
plots are reversed between the two.) It can be observed that the geometric flow patterns of the wind, as represented
by the pressure contours, are generally similarpresenting similar areas of positive and negative (suction) pressure
on the building surface.
Windward surfaces (east face) Area 1: It can be seen in both towers that the front surface exhibits nearly
constant pressure with a slightly lower pressure in the bottom corners. The second tower exhibits the highest
mean pressure near the top while showing a low pressure zone near the left bottom corner of the base.
North face Area 2: Both simulation and wind tunnel results show a zone of low pressure (flow
detachment) at the front of tower 1 and the front of tower 2 that is higher than tower 1 where air has gone
over tower 1, hit tower 2s east face and is going around, creating an area of low pressure on the side as
expected.
Roof surfaces Areas 3 and 4: Tower 1 shows an area of low or negative pressure on the forward edge
and higher pressure toward the rear edge. Pressure on the roof of tower 2 is low relative to the east face of
the tower as revealed by experiment.
Using the pressure data obtained through the validation exercise, a closer comparison can be made between the
simulation and wind tunnel experiments for each of the 511 data points. Results for this specific case are summarized
as follows:
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Validation brief for the Robot Structural Analysis Professional wind simulation tool
It is also observed that on the tower roofs, the simulation tends to over
predict the impacts of air flow over the slanted roofs. As air flows over
the slant of tower 1, the pressure on the leading edge is overestimated
while the lift of the airand the resulting negative pressureon the
back edge of the roof is also overestimated. This same behavior is
observed on the roof of tower 2.
The simulation is reasonably accurate on the mid-height roofs and on
the pedestal building to the right of the towers, but under predicts the
negative pressure on the roof caused by the turbulent air between the
towers.
10
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Figure 8. Wind simulation results for deviation factor of 0.5% (default) and 0.2%.
To answer this, results were examined for 12.89 seconds (the time convergence was reached for DF=0.5%) and 24.3
seconds (the time convergence was reached for DF=0.2% ). The plots below show the results for each duration
compared to experimental results. The following conclusions are made.
WindwardIt can be seen that for the windward face of the structure, the results are quite similar, with the
longer simulation time showing slightly more accurate results in the wake area behind tower 1.
Side FacesAs shown for the north face, the results are quite similar with the shorter simulation showing
more favorable results at lower heights (AAxx through AExx sensors).
Roof SurfacesFor the roof surfaces, results are very similar, trading accuracy in various areas. In general,
the longer simulation appears to reduce the degree to which the simulation overestimates the effect of the
airflow over the slanted roofs as discussed earlier.
11
Validation brief for the Robot Structural Analysis Professional wind simulation tool
LeewardThe leeward face of the structure shows mixed results. Again, the
longer simulation seems to improve results for pressure measured in the
turbulent flow between the two towers as well as the lower area behind
tower 2, suggesting the wake is better simulated with time. The exception
shown by the highlighted segmentsis the area at the edge of the rear slant
of the roof on tower 2. In this area, the results were more accurate earlier in
the simulation.
12
Validation brief for the Robot Structural Analysis Professional wind simulation tool
13
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Conclusions
The Robot Structural Analysis wind simulation tool has been developed to provide building designers with quick and
easy access to wind tunnel simulation, providing insight into air flow behavior and estimates of design loads resulting
from wind. The above examples provide an examination of how the wind simulation tool can perform these functions
in a typical application.
It is observed that the simulation performs well at capturing the geometric flow patterns of the air for moderately
complex building shapes. Quantitatively, the simulation performs reasonably well at estimating the mean surface
pressure resulting from the simulated wind with results giving conservative estimates for pressurespositive and
negativeresulting from flow effects such as detachment (suction) and reattachment.
With this, the wind simulator can prove to be a very useful tool in
providing nearly instant assessments of wind performance in the
early concept and design detail phase of a building project. It can
support code-based analytical analysis in developing design loads
and performing preliminary structural analysis. Finally, the tool can
serve as a practical means of performing trade studies and
test/model optimization before time and money is invested in scalemodel wind tunnel testing that is often required of complex building
projects.
14
Validation brief for the Robot Structural Analysis Professional wind simulation tool
N
E
15
Validation brief for the Robot Structural Analysis Professional wind simulation tool
Figure 10. Layout of pressure sensors on south building elevation (left) and north building elevation (right).
Figure 11. Layout of pressure sensors on east and west building elevation.
Autodesk, the Autodesk logo, and Robot are registered trademarks or trademarks of Autodesk, Inc., and/or its subsidiaries and/or affiliates in the
USA and/or other countries. All other brand names, product names, or trademarks belong to their respective holders. Autodesk reserves the right
to alter product and services offerings, and specifications and pricing at any time without notice, and is not responsible for typographical or
graphical errors that may appear in this document. 2014 Autodesk, Inc. All rights reserved.
16