Dear @Jason.Jonkman,
Hope you are doing well.
I want to ask you whether there is a MATLAB (or Python) function that writes a .bts file ?
Thank you in advance.
Best Regards,
Riad
1 Like
Dear @Riad.Elhamoud
Dear Abhay,
Python script turbsim_file.py in the openFAST-Toolbox reads and writes binary formatted TurbSim (.bts) files: openfast_toolbox/openfast_toolbox/io/turbsim_file.py at main Ā· OpenFAST/openfast_toolbox Ā· GitHub.
From the matlab-toolbox, there is a script readfile_BTS.m that will read a TurbSim binary file into the MATLAB workspace: matlab-toolbox/Utilities at main Ā· OpenFAST/matlab-toolbox Ā· GitHub. Iām not aware that an equivalent MATLAB-based script to write a TurbSim binary file has been developed, but Iām sure that would be possible using readfile_BTS.m as a starting point.
Best regards,
1 Like
Python codes are also available, see here WEIS/weis/aeroelasticse/turbsim_util.py at develop Ā· WISDEM/WEIS Ā· GitHub
1 Like
Iām attempting to generate a BTS file using a user-defined time series file. However, Iāve noticed that in the resulting wind file, both the v and w components are zero. Iām unsure how to proceed.
When I try to use the āIECKAIā TurbModel, it appears that I canāt utilize the time series file. This seems to contradict the comments in the INP file. However, upon reviewing the user manual, I found that time series can only be used with the TIMESR turbulence model.
Iāve attached the relevant files for reference. Here is key information:
This summary file was generated by TurbSim on 13-Aug-2024 at 17:35:04.
Runtime Options:
123456 Random seed #1
RANLUX Type of random number generator
F Output binary HH turbulence parameters?
F Output formatted turbulence parameters?
F Output AeroDyn HH files?
T Output AeroDyn FF files?
F Output BLADED FF files?
F Output tower data?
F Output formatted FF files?
F Output coherent turbulence time step file?
T Clockwise rotation when looking downwind?
0 - NONE IEC turbulence models scaled to exact specified standard deviation
Turbine/Model Specifications:
45 Vertical grid-point matrix dimension
61 Horizontal grid-point matrix dimension
0.500 Time step [seconds]
5000.000 Analysis time [seconds]
5000.000 Usable output time [seconds]
115.000 Hub height [m]
220.000 Grid height [m]
300.000 Grid width [m]
0.000 Vertical flow angle [degrees]
0.000 Horizontal flow angle [degrees]
Meteorological Boundary Conditions:
TIMESR User-input time series spectral model
N/A IEC standard
N/A IEC turbulence characteristic
N/A IEC turbulence type
N/A IEC Extreme Turbulence Model (ETM) "c" parameter [m/s]
PL Wind profile type
90.000 Reference height [m]
10.000 Reference wind speed [m/s]
N/A Jet height [m]
0.130 Power law exponent
0.030 Surface roughness length [m]
Non-IEC Meteorological Boundary Conditions:
45.000 Site latitude [degrees]
0.050 Gradient Richardson number
0.480 Friction or shear velocity [m/s]
N/A Mixing layer depth [m]
N/A Mean hub u'w' Reynolds stress
N/A Mean hub u'v' Reynolds stress
N/A Mean hub v'w' Reynolds stress
Spatial Coherence Models:
IEC u-component coherence model
IEC v-component coherence model
IEC w-component coherence model
( 12.000, 0.659E-03) u-component coherence parameters
( 12.000, 0.659E-03) v-component coherence parameters
( 12.000, 0.659E-03) w-component coherence parameters
N/A Coherence exponent
Profiles from User-Defined Time-Series Input:
Height Wind Speed Horizontal Angle Vertical Angle
(m) (m/s) (deg) (deg)
------ ---------- ---------------- --------------
105.00 12.20 15.00 7.25
105.00 12.14 15.00 7.25
75.00 10.97 15.00 7.25
75.00 10.77 15.00 7.25
You have requested that the following file(s) be generated:
leice.bts (AeroDyn/TurbSim full-field wind file)
Turbulence Simulation Scaling Parameter Summary:
Turbulence model used = User-input time series
Gradient Richardson number = 0.050
Monin-Obukhov (M-O) z/L parameter = 0.067
Monin-Obukhov (M-O) length scale = 1725.000 m
Mean wind speed at hub height = 10.324 m/s
Wind profile type = Power law
Power law exponent = 0.130
Mean shear across rotor disk = 0.020 (m/s)/m
Assumed rotor diameter = 220.000 m
Surface roughness length = 0.030 m
Nyquist frequency of turbulent wind field = 1.000 Hz
Number of time steps in the FFT = 10000
Number of time steps output = 10000
Number of points simulated = 2747
Mean Flow Angles:
Vertical = 0.0 degrees
Horizontal = 0.0 degrees
Mean Wind Speed Profile:
Height Wind Speed Horizontal Angle Vertical Angle U-comp (X) V-comp (Y) W-comp (Z)
(m) (m/s) (degrees) (degrees) (m/s) (m/s) (m/s)
------ ---------- ---------------- -------------- ---------- ---------- ----------
225.0 11.27 0.00 0.00 11.27 0.00 0.00
220.0 11.23 0.00 0.00 11.23 0.00 0.00
215.0 11.20 0.00 0.00 11.20 0.00 0.00
210.0 11.16 0.00 0.00 11.16 0.00 0.00
205.0 11.13 0.00 0.00 11.13 0.00 0.00
200.0 11.09 0.00 0.00 11.09 0.00 0.00
195.0 11.06 0.00 0.00 11.06 0.00 0.00
190.0 11.02 0.00 0.00 11.02 0.00 0.00
185.0 10.98 0.00 0.00 10.98 0.00 0.00
180.0 10.94 0.00 0.00 10.94 0.00 0.00
175.0 10.90 0.00 0.00 10.90 0.00 0.00
170.0 10.86 0.00 0.00 10.86 0.00 0.00
165.0 10.82 0.00 0.00 10.82 0.00 0.00
160.0 10.78 0.00 0.00 10.78 0.00 0.00
155.0 10.73 0.00 0.00 10.73 0.00 0.00
150.0 10.69 0.00 0.00 10.69 0.00 0.00
145.0 10.64 0.00 0.00 10.64 0.00 0.00
140.0 10.59 0.00 0.00 10.59 0.00 0.00
135.0 10.54 0.00 0.00 10.54 0.00 0.00
130.0 10.49 0.00 0.00 10.49 0.00 0.00
125.0 10.44 0.00 0.00 10.44 0.00 0.00
120.0 10.38 0.00 0.00 10.38 0.00 0.00
115.0 10.32 0.00 0.00 10.32 0.00 0.00
110.0 10.26 0.00 0.00 10.26 0.00 0.00
105.0 10.20 0.00 0.00 10.20 0.00 0.00
100.0 10.14 0.00 0.00 10.14 0.00 0.00
95.0 10.07 0.00 0.00 10.07 0.00 0.00
90.0 10.00 0.00 0.00 10.00 0.00 0.00
85.0 9.93 0.00 0.00 9.93 0.00 0.00
80.0 9.85 0.00 0.00 9.85 0.00 0.00
75.0 9.77 0.00 0.00 9.77 0.00 0.00
70.0 9.68 0.00 0.00 9.68 0.00 0.00
65.0 9.59 0.00 0.00 9.59 0.00 0.00
60.0 9.49 0.00 0.00 9.49 0.00 0.00
55.0 9.38 0.00 0.00 9.38 0.00 0.00
50.0 9.26 0.00 0.00 9.26 0.00 0.00
45.0 9.14 0.00 0.00 9.14 0.00 0.00
40.0 9.00 0.00 0.00 9.00 0.00 0.00
35.0 8.84 0.00 0.00 8.84 0.00 0.00
30.0 8.67 0.00 0.00 8.67 0.00 0.00
25.0 8.47 0.00 0.00 8.47 0.00 0.00
20.0 8.22 0.00 0.00 8.22 0.00 0.00
15.0 7.92 0.00 0.00 7.92 0.00 0.00
10.0 7.52 0.00 0.00 7.52 0.00 0.00
5.0 6.87 0.00 0.00 6.87 0.00 0.00
Harvested Random Seeds after Generation of the Random Numbers:
382863125 K 1
0 K 2
Hub-Height Simulated Turbulence Statistical Summary:
Type of Wind Min (m/s) Mean (m/s) Max (m/s) Sigma (m/s) TI (%)
---------------- --------- ---------- --------- ----------- ------
Longitudinal (u) 8.46 10.32 12.12 0.524 5.072
Lateral (v) -0.00 0.00 0.00 0.000 0.000
Vertical (w) -0.00 -0.00 0.00 0.000 0.000
U component 8.46 10.32 12.12 0.524 5.072
V component -0.00 0.00 0.00 0.000 0.000
W component -0.00 -0.00 0.00 0.000 0.000
Horizontal (U&V) 8.46 10.32 12.12 0.524 5.072
Total 8.46 10.32 12.12 0.524 5.072
Min Reynolds Mean Reynolds Max Reynolds Correlation
Product Stress (m/s)^2 Stress (m/s)^2 Stress (m/s)^2 Coefficient
---------------- -------------- -------------- -------------- -----------
u'w' -0.000 -0.000 0.000 -0.004
u'v' -0.000 0.000 0.000 0.002
v'w' -0.000 0.000 0.000 0.028
Friction Velocity (Ustar) = 0.000 m/s
Maximum Instantaneous TKE = 1.740 (m/s)^2
Maximum Instantaneous CTKE = 0.000 (m/s)^2
Any insights or suggestions would be greatly appreciated.
Has anyone else experienced similar issues or know of a workaround? Thank you in advance for your help!
usr-turbsim
Best Regrads,
Junius
Dear @Jundong.Wang,
Iām not sure why the v and w components are showing up as zeros when you specified nonzero values, but I found a couple oddities in your inputs:
- The user-specified points youāve set are far to one side of the grid and the reference point (
RefPtID) is outside the grid altogether - Your user-specified v and w components seem to have a nonzero mean, even though your profile and wind direction are zero.
I would first suggest using one user-specified point, placing this point within the domain, and ensuring that the v and w components have zero mean.
Also, which version of TurbSim are you running?
Best regards,
Dear @Jason.Jonkman ,
Thank you for your reply. I misunderstood the definition of pointyi, thinking it referred to the global coordinate system, but Iāve made the necessary adjustments.
BTW, Turbsim version information:
**************************************************************************************************
TurbSim
Copyright (C) 2021 National Renewable Energy Laboratory
Copyright (C) 2021 Envision Energy USA LTD
This program is licensed under Apache License Version 2.0 and comes with ABSOLUTELY NO WARRANTY.
See the "LICENSE" file distributed with this software for details.
**************************************************************************************************
TurbSim-v2.5.0-648-g2221ba53
Compile Info:
- Compiler: GCC version 9.3.0
- Architecture: 64 bit
- Precision: single
- Date: Apr 6 2021
- Time: 21:20:00
Execution Info:
- Date: 08/15/2024
- Time: 11:21:42+0800
Following your suggestions, I tested three scenarios: single-point data, data from two points at the same height, and data from two points at different heights on the same side. Unfortunately, the velocity components v and w both have zero results.
Iāve uploaded the relevant files to OneDrive for your review. Iām also examining the source code, and Iāll share any updates here.
Dear @Jundong.Wang,
I would first focus on the singe-point test (leice-test1.inp). I still see that the mean value of your user-specified v and w components have a sizeable nonzero mean. Do the results make more sense if you remove this mean?
Also, does upgrading to the newest version of TurbSim provided with OpenFAST resolve the issue: https://github.com/OpenFAST/openfast/releases/download/v3.5.3/TurbSim_x64.exe.
Best regards,
@Jason.Jonkman ,
Thank you so much for your suggestion. Iāve been using the latest version of TurbSim for the simulation. I apologize for only just noticing the issue you mentioned about the non-zero mean of the v and w velocity components. I suspect the reason the simulation results were zero is that the wind speed data I provided has significant values in both the horizontal and vertical directions, whereas I intended for the data to generate as zero. However, it seems there are also quite a bit of fluctuations in the reference documents provided by TurbSim. Iām unsure how the average inflow angle is set.
Once I removed the time average from the v and w components, the program ran properly. You can check it out from the following link:
3 items
Thanks to your kind reminder.
I may need to further confirm if I should adjust the average inflow angle at different height layers by modifying the source code to better match the data.
Best Regards,
Junius
Dear @Jundong.Wang,
Iām glad that solved the problem. TurbSim generally works by superimposing turbulent fluctuations with zero mean with a mean wind profile. The mean wind profile can include a change of wind speed with height (shear) as well as a change in wind direction with height (veer), depending on what profile youāve selected.
Best regards,