Using the output from one OLAF simulation as the input for another

[justhawk98]

I am curious how well OLAF can model the wake of a waked turbine. I am aware OLAF can't be used with FAST.Farm but I had an idea. Since I am only interested in a two turbine case, one turbine operating in the wake of another, I was wondering if I could record the wake on a yz plane, then use that recorded wake as the input to a 2nd OLAF simulation. I have a working OLAF simulation of one turbine operating in steady inflow where the wake is output in the vtk format at every time step. I'm just wondering if there is a way to take those vtk files and convert them to an input for a second OLAF simulation. If anyone has any suggestions or could point me in the right direction, please let me know.

[jjonkman]

Dear @justhawk98,

I would say OLAF computes the wake well in the near wake (< 3-4D), but does not work well in the far wake( > 3-4D), where the concentrated vortex structures in the wake break down and what remains is meandering wake deficit and wake-added turbulence, which is what FAST.Farm models well. Ideally, OLAF would be combined with FAST.Farm so that both the near- and far-wake are captured accurately, with the near wake being important for the aerodynamic loading of the rotor generating the wake and the far wake being important for the impact of that wake on a downstream turbine. In fact, this is future development work that we are planning.

The big question is how far your downstream turbine is spaced relative to your upstream turbine? If the spacing is close, OLAF is preferred and if the spacing is far, FAST.Farm is preferred.

FYI: OLAF can already model (closely spaced) multi-rotor system through the standalone AeroDyn driver and we are working to support multi-rotor capability directly within OpenFAST (including aero-hydro-servo-elastics).

Best regards,

[justhawk98]

Thank you for the info. I'm looking at a 3-9 diameter spacing between the turbines. Though, the bigger picture here is I'm trying to compare the performance of AMR-Wind, OLAF, and a code our lab created here at BYU. I can set up a single turbine comparison pretty easily but I also wanted to compare the behavior of two turbines interacting. So I was just looking for some kind of workaround to feed OLAF the wake of another turbine as its input, even if this might not be the intended use case. I would be interested in the multi-rotor OLAF simulation through AeroDyn, if you think that would work better here.

[jjonkman]

Dear @justhawk98,

The mulitirotor capability of the standalone AeroDyn driver is useful for running multiple instances of OLAF together, as long as you are OK with rigid turbines and prescribed motion rather than aero-elastic interaction. But, again, OLAF is best suitable for the near wake, not the far wake, so maybe OLAF is not best option for your use case.

Best regards,

[justhawk98]

Sounds perfect for what I'm doing. Thank you!

[justhawk98]

Here's the result. I was able to get two turbines running in the Aerodyn driver. Both wakes simulated via OLAF. The above image is an x-y plane slice at hub height of the two turbines. They are aligned in the direction of the wind and separated by a spacing of 3 diameters. This is plenty for my use case. Thanks for your help!