Dear wind turbine enthusiast,
TURBSIM is a stochastic wind speed field simulator where it generates wind speed fields taking into account spatial and temporal variability right ?
Why there is no spatial variability in sea elevation field ? or the spatial variability in sea elevation field is somehow to do with the spreading wave range in the short-crested wave ?
Another question: what is the meaning of “the wind is simulated using 10-minute recurring time histories” ? Can TURBSIM generate this kind of wind speed time history ? If yes, then how to do it ?
Thanks in advance.
Best Regards,
Riad
Dear @Riad.Elhamoud,
Regarding spatial and temporal variability, the temporal variability in TurbSim comes from the underlying spectrum and phases of the frequency compnents, and SeaState uses a similar approach to generate temporal variability of its irregular sea state. The spatial variability in TurbSim comes through the underlying spatial coherence based on exponential coherence functions (or use-defined), whereas in SeaState comes through use of the dispersion relationship.
Regarding “recurring time histories”, I’m not sure the source of your quote, but perhaps you are referring to fully periodic time histories? Fully periodic wind data can be output from TurbSim by setting UsableTime = “ALL”, which then allows you to run simulations in OpenFAST that are longer than TurbSim’s AnalysisTime, with the data repeating with a period of AnalysisTime.
Best regards,
1 Like
Thank you @Jason.Jonkman for your reply.
Yeah, you are correct, i mean fully periodic time histories using TURBSIM.
I would like please to hear your opinion (whether you agree or not) on the following statement:
“When simulating a FOWT for 3 hours, fully periodic time histories generated by TURBSIM are used for an analysis period of 10 minutes, as the wind speed is assumed to be stationary over this 10-minute interval. In contrast, the wave elevation is simulated over the full 3 hours, since it is considered stationary only over that longer period”.
Thanks in advance,
Best Regards,
Riad
Dear @Riad.Elhamoud,
You could certainly use OpenFAST to simulate a 3-hour seastate with repeated 10-minute periodic winds as you outline, but there are other ways to approach the problem of different stationary periods between winds and waves. We studied this problem a bit in publication 12 years ago–see: https://docs.nrel.gov/docs/fy13osti/58153.pdf and https://docs.nrel.gov/docs/fy13osti/58518.pdf.
Best regards,
1 Like
Dear @Jason.Jonkman ,
Sorry for re-opening this topic.
For the wind speed field: only temporal and spatial variability along-y occur. No variability along x-direction (direction of propagation of the incodent wind) and no variability along z-direction right ?
For the wave elevation field: temporal and spatial variability (along x and y) occur. No variability along z-direction since we are dealing with elevation right ?
Is it possible to visualize the stochastic wave elevation field generated by SeaState ? According to your knowledge, does exist a functin in the MATLAB toolbox like that for the stochastic wind speed field named :Read_BTS.m ?
Best Regards,
Riad
Dear @Riad.Elhamoud,
For wind, a TurbSim wind field is a 2D in space (Y,Z) and 1D in time. Taylor’s frozen turbulence assumption in X/time is used by InflowWind to generate a 3D (X,Y,Z) + time wind field. Turbulence does vary along Z from spatial coherence and the wind profile.
For waves, SeaState generates a 3D in space (X,Y,Z) + time wave field directly. The wave kinematics vary in Z (hyberbolic functions that decrease the wave kinematics with depth). The wave elevation field is 2D in space (X,Y) + time.
Best regards,
1 Like