Dear ALL,
Does anyone have the FAST model for SWAY wind turbine accommodating to the following documents ?
nrel.gov/docs/fy15osti/63569.pdf
nrel.gov/docs/fy16osti/61744.pdf
since I can’t find the first author.
Thanks a lot.
Kyushu University
Yingyi Liu
Dear Dr. Jason,
Do you have an archive for the SWAY wind turbine FAST model?
We would like to study the aerodynamic lift forces on its tower.
If you have, could you upload it?
Thanks a lot.
best regards,
Yingyi Liu
Dear Yingyi Liu,
We were very pleased with how much information SWAY allowed us to publish, but because of the proprietary nature of the SWAY system, we are unable to share the FAST input files that were used in the reports you references. Much of the information needed to build a FAST model of the SWAY system is available in those reports if you wish to build a FAST model yourself using the publicly available information.
Best regards,
Dear Dr.Jason/Bonnie,
Thank you for your reply. Based on your information, we decide not to use the SWAY model since it is proprietary and our target is just realization of an arbitrary airfoil-shaped tower in FAST computation. To simply give an example, we temporarily use the NACA0012 airfoil for the tower section shape.
And currently, I have already finished modifying the latest FAST_v8 code to enable its computation of aerodynamic tower lift force and pitching moment, together with the existing tower drag force, on an arbitrary airfoil-shaped tower. However, the calculated aerodynamic tower forces and moment have not mapped to the structural part yet, which means, the code is now only able to compute the forces and moment and then simply output.
But I feel difficulty in finding the subroutines that map the tower aerodynamic forces and moment to the code of structural part in FAST_v8. For the aerodynamic forces, it may be realized by simply changing the drag force to a combination of drag and lift. But for the pitching moment, how should I modify the existing code in mapping to the code of structural part? Are there many places that need to be modified?
Thank you and best wishes,
Yingyi Liu
Dear Yingyi Liu,
This should be quite straightforward, as the mesh-mapping routines already implemented in FAST do the work for you. Presumably you’ve added your tower lift and pitching moment loads to the AeroDyn v15 output y%TowerLoad. This output is created with both force and moment fields, which have already been mapped as input forces and moments to the ElastoDyn tower model. So, there is nothing that needs to be changed outside of specifying the tower lift and pitching moment loads to AeroDyn v15.
Best regards,
Dear Dr. Jason,
Thank you for your reply. We have implemented it according to your suggestion. I have another question, though.
Fx=Fdcos(alpha)-Flsin(alpha)
and
Fy=Fdsin(alpha)+Flcos(alpha)
respectively. Well, this definition conflicts with the Hansen’s definition in his book (“Aerodynamics of wind turbine”, second edition, M.O.L. Hansen, 2008, page 48, see below attached screenshot) that
Fx=Flsin(alpha)-Fdcos(alpha) (PT, Eq.6.11)
and
Fy=Flcos(alpha)+Fdsin(alpha), (PN, Eq.6.10)
with a difference of sign in the Fx component.
In addition, the direction of pitching moment also has something strange. In the present first figure, the positive pitching moment direction is pointing downward along negative z direction (based on the right-hand rule) if viewed from the tower top to the ground level (up-to-down).
I am not sure how the coordinate system of tower section is defined (or by default) in FAST. In order to make sense of it, should I change the positive direction of the y-axis (or x-axis ?) to the opposite, or change the normal direction of the pitching moment to make it point upward along the positive z direction?
I would very much like to know your opinions.
Thank you and best wishes,
Yingyi Liu
Dear Yingyi,
I’m not sure it matters how x, y, and z are located in the airfoil, as long as you are consistent in the coordinate transformations.
The local tower coordinate system in AeroDyn v15 is shown in Figure 2 of the draft AeroDyn v15 User’s Guide and Theory Manual: wind.nrel.gov/nwtc/docs/AeroDyn_Manual.pdf.
Regarding the pitching moment, normally, a positive pitching moment implies a load that will push the nose up, increasing the angle of attack.
I think part of the problem in your figure is that usually the cross-sectional image of an airfoil is shown looking from the root of the blade toward the tip. For the tower, this would be the same as looking from the tower base up to the top. However, your image shows you looking from the tower top down to the base.
I hope that helps.
Best regards,
Dear Dr.Jason,
Thanks for your comments which are very helpful. We have modified our implementation based on your suggestions.
We plan to run the modified Aerodyn_v15 tool together with Adams to assist the analysis. Unfortunately, we found that FAT_v8 does not have the capability of generation of an FAST2ADMAS model. Therefore, unluckily, the current work seems could not proceed.
Under such a situation, we would be very eager to know that, do you have a schedule of implementing the FAST2ADMAS capability in FAST_v8 for the future plans? If so, would you mind indicating an approximate date when it will be released? If not, is it easy for us to implement this capability in FAST_v8 by our own self, could you provide several general guidance? Since we would be very pleased to release our implementation to the public if it can be successfully completed.
Best regards,
Yingyi Liu
Dear Yingyi Liu,
We have no plans at NREL to introduce the FAST-to-ADAMS functionality into FAST v8.
You could develop this capability yourself, but I expect that this would take a significant effort. I suggest that you start by reviewing how the FAST-to-ADAMS functionality of FAST v7 worked. You will also need to update the ADAMS-to-AeroDyn (A2AD) interface to be compatible with the modules you need (AeroDyn v15, ServoDyn, HydroDyn, etc.).
If you want to use MSC.ADAMS, is there a reason why you don’t just use the FAST-to-ADAMS functionality of FAST v7 and corresponding A2AD interface?
Best regards,
Dear Dr. Jason,
Thank you for your answer. We didn’t use FAST_v7 because there isn’t a module for calculating tower drag force in it. As we find FAST_v8 has this module, we started work on modifying the FAST_v8. However, soon we find FAST_v8 doesn’t have FAST2ADAMS functionality. Since as you mentioned, implementation of FAST2ADAMS functionality into FAST_v8 take a significant effort, and currently due to the project tight schedule, we seem not have enough time to do this.
A best way to solve this may be, that I could work directly on a version of FAST_v7 that has the module of calculating tower drag force and has already been incorporated with the structural part of the FAST_v7 code. Then I could do immediately the same work I have done by adding the functionality of lift and pitching moment, which saves time. So Dr. Jason, I would like to know that, do you have such a version at hand for being released? since I know that you have been working on this issue before by a collaboration on the SWAY wind turbine.
Best Regards,
Yingyi Liu
Dear Yingyi Liu,
NREL never released a version of FAST v7 with aerodynamic drag loads on the tower. Depending on the level of sophistication you need in this calculation, it may not be hard to simply added loads to your ADAMS model after the model is generated via the FAST-to-ADAMS functionality.
Best regards,
Dear Dr. Jason,
I would like to know that (1) Is it difficult to add the functionality of tower aerodynamic drag, lift forces and pitching moment to the existing FAST v7 code?
(2) Further on, if that were completed, can it be able to export the corresponding ADAMS model involving the tower aerodynamic loads using the existing build-in FAST2ADAMS functionality?
(3) Can simply adding loads to ADAMS model have sufficient accurate results, in similarity to the direct modification on FAST v7 code?
Thank you for your reply in advance.
Best regards,
Yingyi Liu
Dear Yingyi
The level of difficulty to add tower aerodynamic loads to FAST v7 likely depends on how sophisticated you want to make it. Adding the functionality to FAST v7 does not mean that it automatically transfers to FAST-to-ADAMS; you’ll have to modify FAST-to-ADAMS and the ADAMS-to-AeroDyn interface as well. I’m not sure I understand your third question, but again, the answer is likely it depends on how sophisticated you want to make it.
Best regards,
Dear Dr. Jason,
Thank you for your answer. About the third question, I mean that if we add simply the tower aerodynamic loads to your ADAMS model after the model is generated via the FAST-to-ADAMS functionality, are these newly added loads constant with time? Because during the calculation, the program (either FAST or ADAMS) needs to look up the tower aero data from external files at each time step, it might be necessary to apply non-constant aerodynamic forces to the tower. I would kindly ask for your suggestion: from your experience in using ADAMS, do you think it is possible to do this in such a way, by looking up the tower aero data from external files and calculating the tower aerodynamic loads at each time step in ADAMS?
Thank you so much.
Yingyi Liu
Dear Yingyi,
It would be quite simple to add constant tower aerodynamic loads to an existing ADAMS model. With a bit more work, you could add time-varying aerodynamic loads. To read in wind or aerodynamic loads from a file could also be done, but will take further effort.
Best regards,
