Change Turbine Type #1132
-
|
Hello, I am new to FLORIS and have a question regarding its usage. For helix active wake mixing, I'm currently using the example input file "emgauss_helix.yaml", which by default uses the iea_15MW turbine model. If I want to apply helix active wake mixing to the nrel_5MW turbine instead, what parameters need to be changed? Is it sufficient to simply replace the turbine type from iea_15MW to nrel_5MW, or are there additional parameters that should also be adjusted to ensure correct performance? Thank you in advance for your help! Best regards, |
Beta Was this translation helpful? Give feedback.
Replies: 3 comments
-
|
Hi @niuyue321 , Thanks for the question! The Helix active wake mixing model has several parameters to describe how applying the Helix with a certain amplitude affects the power and thrust curve, described here. However, at this point, those parameters have only been calibrated for the IEA 15MW turbine, and it's likely they wouldn't be directly transferable to the NREL 5MW. I'm tagging @jfrederik-nrel as he was the one that developed the AWC turbine model and might have some insights on whether the parameters for the IEA 15MW could reasonably be used also for the NREL 5MW (or what issues might arise if doing this). Misha |
Beta Was this translation helpful? Give feedback.
-
|
Hi @niuyue321, Everything @misi9170 said is correct. In terms of power loss of the turbine applying the helix, I would not expect substantial differences between the IEA 15MW and the NREL 5MW turbine. However, I would still recommend recalibrating these parameters using e.g. OpenFAST simulations using the AWC_mode in ROSCO, see this example. In terms of the additional turbulent mixing induced by the helix, and subsequently, the power uplift of downstream turbines, the default parameters were tuned to match specific turbulent LES data. Just like with all results obtained from FLORIS, the reliability heavily depends on the accuracy of your tuning process. You can certainly use the current values--it should run without errors--and see what you get for the NREL 5MW turbine, but results should be taken with a large grain of salt. If you don't have access to tuning data from higher-fidelity simulations, I would recommend sticking to the IEA 15MW turbine to get somewhat reliable results, but the code should also work for the NREL 5MW turbine. If you do end up running FLORIS using the NREL 5MW turbine with the current tuning parameters, I'd be curious to see the results, and would be happy to provide feedback on whether the results looks reasonable or not. Good luck, Joeri |
Beta Was this translation helpful? Give feedback.
-
|
Dear Misha @misi9170 and Joeri @jfrederik-nrel, Thank you both so much for your detailed and helpful responses. I truly appreciate your help! Best wishes, Yue |
Beta Was this translation helpful? Give feedback.
Hi @niuyue321,
Everything @misi9170 said is correct. In terms of power loss of the turbine applying the helix, I would not expect substantial differences between the IEA 15MW and the NREL 5MW turbine. However, I would still recommend recalibrating these parameters using e.g. OpenFAST simulations using the AWC_mode in ROSCO, see this example.
In terms of the additional turbulent mixing induced by the helix, and subsequently, the power uplift of downstream turbines, the default parameters were tuned to match specific turbulent LES data. Just like with all results obtained from FLORIS, the reliability heavily depends on the accuracy of your tuning process. You can certainly use the current v…