Dear @Muhammad.Sharjil,
The tower potential flow model will distort the flow around the tower, which can impact the performance of the rotor a bit. Often the tower influence is neglected when computing steady state rotor performance, but it certainly can be included if it has a sizeable influence.
To azimuth-average the values, simply compute the mean over one revolution of the rotor (or the mean over many revolutions if cycles are “noisy”.).
Best regards,
Hi @Jason.Jonkman
Thanks for the detailed response. Can you give a rough guess in comparison to steady state response, regarding when to consider TwrPotent as significant?
Best Regards,
Muhammad Sharjil
Dear @Muhammad.Sharjil,
I’m not sure I fully understand your question, but the importance of tower influence on rotor performance will depend on the tower diameter, as well as on the rotor overhang, tilt, precone, and blade deflection.
Best regards,
Hi @Jason.Jonkman
I wanted an expert opinion regarding how can we term “significant Tower influence” based on the Cp response in OpenFast i.e, just to have a rule of thumb. The change in above Cp plot is around 1%.
I have another query.
I used Aerodyn_Driver to get Cp, Ct and Cq surfaces against TSR and Blade Pitch Angle.
However, I am getting different values in comparison to
a. Control Design : Page 28 https://www.nrel.gov/docs/fy20osti/76773.pdf
b. IEA-15-240-RWT/OpenFAST/IEA-15-240-RWT/Cp_Ct_Cq.IEA15MW.txt at master · IEAWindTask37/IEA-15-240-RWT · GitHub
Cp value comparison at, TSR = 9, with Blade Pitch Angle = 0,
(for a), Cp = 0.480
(for b), Cp = 0.4692
Computed from Aerodyn_Driver, Cp = 0.4871
(Moreover, this Cp value at TSR = 9 & Blade Pitch Angle = 0, is maximum for the Cp data I have generated with Aerodyn_Driver.)
ad_drvr
aerodyn
Kindly guide me on this.
Thanks
Dear @Muhammad.Sharjil,
I’ll leave it to you define what is a “significant” change in Cp is to you.
I was not the one to develop the Cp surface for the IEA Wind 15-MW reference wind turbine, so, I’m not sure how your set up differs.
Best regards,
Hi @Jason.Jonkman
Thanks for the response.
Regards,
Muhammad Sharjil
hi @Jason.Jonkman
I am facing difficulty in getting linear model for IEA 15MW with UMaineSemi on OpenFast 3.4.1
DoF Enabled
a. Drivetrain rotational-flexibility DOF
b. Generator DOF
All other DoF are disabled
Operating Condition:
RPM = 7.55
Wind Speed = 16 m/sec
I am getting the following error
Elastodyn
Servodyn
Thanks
Regards,
Muhammad Sharjil
Dear @Muhammad.Sharjil,
When linearizing OpenFAST in above-rated conditions, we typically recommend setting constant generator torque in ServoDyn and trimming blade pitch. Are you using TrimCase = 3 in the OpenFAST primary input file? I don’t see that you are using a constant generator torque in ServoDyn. With VSContrl = 1, you should set:
VS_RtGnSp = some small number greater than zero, e.g., 0.001
VS_RtTq = the desired generator torque (around 19.6 MNm as you’ve set for the IEA Wind 15-MW RWT)
VS_Rgn2K = some small number greater than zero, e.g., 0.001
VS_SlPc = some small number greater than zero, e.g., 0.001
Best regards,
Hi @Jason.Jonkman
I am using TrimCase = 3 in .fst file.
I tried the suggested numbers in Servodyn. I am still getting the same errors.
Can you guide me on this.
Thanks
Best Regards,
Muhammad Sharjil
Dear @Muhammad.Sharjil,
Presumably you are starting with the OpenFAST model of the IEA Wind 15-MW RWT atop the UMaine semisubmersible from the IEA Wind Task 37 repository (IEA-15-240-RWT/OpenFAST at master · IEAWindTask37/IEA-15-240-RWT · GitHub); is that correct? I see that you’ve enabled the drivetrain DOF of this model (DrTrDOF = TRUE), which was not enabled in the original model. Did you set realistic properties of the drivetrain or reduce the times step (DT) of this model? If not, I suspect that this is causing the model to be numerically unstable and that disabling DrTrDOF will resolve the issue.
Best regards,
Hi @Jason.Jonkman
Thanks for highlighting the differences.
I now used time step of 0.005 sec (instead of default 0.025 sec) and was able to get the linear model for drivetrain DOF + Generator DOF.
Regarding the numbers setting for VSContrl = 1,
I have run the linearization routine with the default gains as provided in IEA-15-240-RWT/OpenFAST at master · IEAWindTask37/IEA-15-240-RWT · GitHub
and
also the numbers you suggested i.e.
VS_RtGnSp = some small number greater than zero, e.g., 0.001
VS_RtTq = the desired generator torque (around 19.6 MNm as you’ve set for the IEA Wind 15-MW RWT)
VS_Rgn2K = some small number greater than zero, e.g., 0.001
VS_SlPc = some small number greater than zero, e.g., 0.001
I compared linear model for below mentioned 2 cases. (3 dof enabled)
I am getting somewhat different linear model (dc gain and Drivetrain mode damping in Transfer Function[Generator RPM / Blade Pitch Angle])
a. Default Gains in Servodyn for Simple Variable Speed Torque Control
b. Suggested Gains in Servodyn for Simple Variable Speed Torque Control
Any comment on which one is depicting the dynamics at this operating point?
Thanks
Regards
Muhammad Sharjil
Dear @Muhammad.Sharjil,
When linearizing above rated, we typically recommend linearizing with constant generator torque set via VSContrl because the real set point would be at the “knee” of the torque-speed curve (between Regions 2.5 and 3) of the simple variable-speed torque control model, which cannot be reasonably linearized.
Best regards,
hi @Jason.Jonkman
Thanks a lot.
Regards,
Muhammad Sharjil
