Confusions about Compiling FAST

Hi all,

I am trying to do some research on the wind farm. Firstly, I want to do research on a specific 3.6MW wind turbine. If I have all the data about the specific 3.6MW wind turbine, can I change the parameters of the NREL 5MW wind turbine into the wind turbine I wanted? Does compiling FAST realize this? Secondly, is it possible to build a wind farm using the single NREL 5MW wind turbine? Is there any good software for doing wind farm research? Thanks a lot for your kind help!

All the best.

Yanhua

Dear Yanhua,

The input parameters available within the various FAST input files will allow you to define a wind turbine model of any size (likely without changing the source code). You shouldn’t need to recompile FAST unless you want to make changes to the source code.

What type of wind farm software do you require? NREL has developed the low-fidelity FLORIS tool (nwtc.nrel.gov/SOWFA), and the soon-to-be released FAST.Farm tool (discussed several times on this forum).

Best regards,

Dear Jason,

Thanks for your timely answer. If I want to define a wind turbine model of 3.6MW, should I need to change the input parameters in Test18.fst and the related ElastoDyn.dat, BeamDyn.dat,InflowWind_12mps.dat,AeroDyn15.dat,ServoDyn.dat? Actually, I want to do some research about designing control method to mitigate the wake effect of the different wind turbines in a wind farm give that I have obtained the wake model. So I will concentrate on wind farm control. Can you tell me which is better for me, FLORIS, SOWFA or the soon-to-be-released FAST.Farm tool, please? Thanks a lot.

All the best.

Yanhua

Dear Yanhua Liu,

Yes, you must modify the various FAST input files in order to create a model that NREL has not already made available e.g. a 3.6-MW wind turbine. Of course, you’ll need many detailed properties e.g. aerodynamics, structures, control etc. to create such a model.

FLORIS has a steady-state solution so it is useful for calculating mean quantities like wind-farm power performance, but cannot predict dynamic quantities like structural loads. FLORIS is often used for wind-farm controls research involving e.g. wake steering, induction control, wind turbine location optimization etc.

SOWFA has a dynamic solution to resolve much of the wind-farm and wind turbine physics, but is computationally expensive, so, you’ll need a large computational resource and you are limited in the number of simulations you can run–typically a handful of simulations for any given problem. SOWFA is often used to understand the underlying physics and to spot check the effects not captured by simpler models.

FAST.Farm tries to mimic SOWFA, but based on some simplifying assumptions so as to enable computational speed-up of the solution. The main use for FAST.Farm is to compute structural loads of the wind turbines in a wind farm through the running of many load-case simulations (similar to how FAST is used to model an individual wind turbine).

Best regards,

Dear Jason,

Thanks a lot. It seems like SOWFA are better to me before FAST.Farm releases. I still have some confusions. Does SOWFA provide outputs of structural loads? Is it possible for us to change the individual wind turbine parameters in SOWFA? Does FLORIS, SOWFA or FAST.Farm provide interface to Simulink? If not, is it possible to design any control methods in SOWFA?

All the best.

Yanhua

Dear Yanhua,

SOWFA may be optionally interfaced to FAST to enable the calculation of wind turbine structural loads. The wind turbine is defined by the FAST model and how the rotor aerodynamics are calculated by SOWFA e.g. actuator disk or actuator line (the latter is required when FAST is interfaced to SOWFA).

None of these software currently have an interface to Simulink. In SOWFA, controls are implemented in FAST for each wind turbine (by subroutines or a dynamic library) ,as well as a wind-farm-wide supercontroller written as a dynamic library that is linked to SOWFA. See the SOWFA documentation for more information.

Best regards,

Dear Jason,

Thanks a lot. It is much more clear to me.

All the best.

Yanhua