Add unsteady aerodynamic model for turbine tail fin #1874

abhineet-gupta · 2023-11-15T19:59:43Z

READY TO MERGE.

ToDo before pull request is ready

Do existing tests pass (Same as the dev branch)
Update existing r-test for tail fin
Add a new r-test for the tail fin USB based on the validation exercise
Address review comments

Feature or improvement description
Implement an unsteady aerodynamic model for tail fin of turbines.

Impacted areas of the software
AeroDyn

Additional supporting information

Tail fin input file changed to obtain additional user-defined data
No changes in the structure of data passed to and from AeroDyn

Test results, if applicable
In future, add the distributed wind reference wind turbine being developed in r-test and add tests for tail fin aerodynamics

ebranlard

I have some small suggestions for the documentation and code, but overall approve this pull request. Thanks a lot!

Please note that the tailfin r-test for the polar based does not seem to pass (likely due to the change of input file).

We would also need to add another r-test for the USB (one of your validation case for instance). Please make sure the unit tests runs quicly (e.g. around 0.5s) and generate minimal but useful data (e.g. few kb). Consider adding a README.md in your rtest folder to describe the system being used/tested.

docs/source/user/aerodyn/input.rst

docs/source/user/aerodyn/theory_tailfin.rst

modules/aerodyn/src/AeroDyn.f90

…Dyn.f90

abhineet-gupta · 2024-02-15T22:49:45Z

The documentation updates can be previewed here:

Section 4.2.1.9.3 of TailUnsteadyTheory.pdf
Section 4.2.1.3.5 of TailUnsteadyInput.pdf

…odel

andrew-platt · 2024-03-08T19:21:44Z

All issues from @ebranlard's review have been addressed.

jjonkman

Sorry for the delay. I see that this has been merged already, but I finally had a chance to review. Please find my comments below.

jjonkman · 2024-04-02T21:29:08Z

docs/source/user/aerodyn/input.rst

@@ -883,22 +883,26 @@ An example of tail fin input file is given below:
    0         TFinIndMod  - Model for induced velocity calculation {0: none, 1:rotor-average} (switch)
    ====== Polar-based model ================================ [used only when TFinMod=1] 
    1        TFinAFID - Index of Tail fin airfoil number [1 to NumAFfiles]
-    ====== Unsteady slender body model  ===================== [used only when TFinMod=2] 
-    [TODO inputs for model 2]
+    ====== Unsteady slender body model  ===================== [used only when TFinMod=2]


Suggest adding units after each variable.

For unitless variables, I would use (-). According to the theory/code, it looks like TFinSigma has the units of 1/degrees.

jjonkman · 2024-04-02T21:29:58Z

docs/source/user/aerodyn/input.rst

-    ====== Unsteady slender body model  ===================== [used only when TFinMod=2] 
-    [TODO inputs for model 2]
+    ====== Unsteady slender body model  ===================== [used only when TFinMod=2]
+    0.9           TFinKp        - Tail fin moment of area about reference point


This comment for TFinKp is not correct. TFinKp is the potential lift coefficient.

jjonkman · 2024-04-02T21:33:43Z

docs/source/user/aerodyn/theory_tailfin.rst



+where :math:`\sigma_i` are empirical constants characterizing the decay of separation functions, :math:`\gamma_{tf}` is the yaw angle of the tail fin with respect to the free-stream wind (:math:`V_{\text{wind}}`), :math:`\alpha^*_i` are the characteristics angles for separation functions.
+:math:`x_i` takes on a value between 0 and 1, and are used to activate or deactivate a the contribution of potential lift, vortex lift and draft to the normal force on the tail fin.


Change "draft" to "drag here.

jjonkman · 2024-04-02T21:36:30Z

modules/aerodyn/src/AeroDyn.f90

+      ! Unsteady aerodynamic model
+
+      ! Calculate separation function (quasi-steady)
+      x1 = 1.0_Reki/(1.0_Reki+exp(p%TFin%TFinSigma(1)*((ABS(gamma_tf)*180.0_ReKi/pi)-p%TFin%TFinAStar(1)))) 


Could use NWTC Library parameter R2D instead of 180/pi here.

jjonkman · 2024-04-02T21:43:56Z

docs/source/user/aerodyn/input.rst

@@ -945,7 +949,21 @@ between 1 and ``NumAFfiles`` and is only used when TFinMod is set to 1. (-)

 Unsteady slender body (USB) model inputs
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+Refer to :numref:`TF-aerotheory` and :cite:`ad-hammam_NREL:2023` for guidance on how to select parameters for the unsteady slender body theory based model.


Wasn't there supposed to be preprocessor developed by Hammam et al so that these parameters could be set directly based on the shape of the tail fin (delta, rectangular, elliptic, etc.)? Was this ever developed? If so, could we share a link?

jjonkman · 2024-04-02T21:57:35Z

docs/source/user/aerodyn/theory_tailfin.rst

@@ -163,9 +161,25 @@ The user only needs to indicate the index `TFinAFIndex` within the list `AFNames
 Unsteady slender body model
 ---------------------------

-The unsteady slender body (USB) model is documented in :cite:`ad-hammam2022`.
+The unsteady aerodynamics of the tail fin is modeled based on Unsteady Slender Body Theory.


Suggest summarizing the limitations of the implementation here, i.e., tail fin chord assumed much smaller than tail fin arm length, and quasi-steady conditions (because your separation functions are implemented only quasi-statically). Perhaps there are others stated in Hammam et al.

jjonkman · 2024-04-02T22:01:13Z

docs/source/user/aerodyn/theory_tailfin.rst

+
+.. math::  :label: TFUSBForce
+
+    N = \frac{\rho}{2} A_{tf} \bigg(  K_p x_1 V_{\text{rel},x} V_{\text{rel},y} +  \Big[x_2 K_v+(1- x_3)C_{Dc} \Big] V_{\text{rel},y}\big|V_{\text{rel},y}\big|\bigg)


There is a negative sign out front in Hammam et al. I presume that is not needed here due to the sign convention used in AeroDyn? I haven't confirmed this, but it may be good to state that.

Presumably this implementation matches results from Hammam et al?

jjonkman · 2024-04-02T23:03:16Z

docs/source/user/aerodyn/theory_tailfin.rst


-The theory will be implemented and documented in a future release.
+    x_i = (1+exp{[\sigma_i (|\gamma_{tf}|-\alpha^*_i)]})^{-1}


I don't see an ABS() in Hammam et al. It makes sense to me that this would be needed, but I suggest mentioning that difference.

jjonkman · 2024-04-02T23:03:43Z

docs/source/user/aerodyn/theory_tailfin.rst



+where :math:`\sigma_i` are empirical constants characterizing the decay of separation functions, :math:`\gamma_{tf}` is the yaw angle of the tail fin with respect to the free-stream wind (:math:`V_{\text{wind}}`), :math:`\alpha^*_i` are the characteristics angles for separation functions.


Why are the separation functions computed using ambient wind while the force calculation uses relative wind (including induction)? Shouldn't the same velocities (which I assume should be the relative velocity) be used for all?

Tail fin aerodynamics: address changes requested in pull request #1874

andrew-platt force-pushed the dev branch from ab957ea to 92c8754 Compare November 29, 2023 17:48

abhineet-gupta changed the base branch from dev to rc-3.5.2 December 7, 2023 17:09

abhineet-gupta changed the base branch from rc-3.5.2 to dev December 7, 2023 17:09

abhineet-gupta added 2 commits December 7, 2023 10:22

Add unsteady aerodynamic model for turbine tail fin

8719168

Add modify tail fin USB model to use gamma instead of alpha

128ff34

abhineet-gupta force-pushed the f/tail_fin_unsteadymodel branch from 562de8b to 128ff34 Compare December 7, 2023 20:56

abhineet-gupta added 2 commits December 7, 2023 14:01

Fixed reg-test link

0bd0c50

Add documentation about tail fin unsteady model

af30fd8

abhineet-gupta marked this pull request as ready for review December 11, 2023 17:41

ebranlard self-requested a review December 11, 2023 18:10

ebranlard suggested changes Dec 12, 2023

View reviewed changes

andrew-platt self-assigned this Jan 10, 2024

abhineet-gupta added 2 commits January 31, 2024 17:31

Address comments on PR 1874 by Emmanuel Branlard

a261f56

TailFinUnsteadyAero: Add regression tests and fix a small bug in Aero…

6e5f71b

…Dyn.f90

andrew-platt added the Module: AeroDyn label Feb 15, 2024

andrew-platt added this to the v4.0.0 milestone Feb 15, 2024

andrew-platt requested a review from jjonkman March 7, 2024 20:18

andrew-platt added 2 commits March 8, 2024 12:17

Merge remote-tracking branch 'OpenFAST/dev' into f/tail_fin_unsteadym…

d709405

…odel

AD: code alignment (noted in PR)

1b3f797

Update input file for Tailfin_FreeYaw1DOF_Unsteady

9725647

andrew-platt merged commit 6a419c7 into OpenFAST:dev Mar 11, 2024
21 checks passed

jjonkman reviewed Apr 2, 2024

View reviewed changes

abhineet-gupta added a commit to abhineet-gupta/AG_openfast that referenced this pull request Aug 16, 2024

AD: Address comments on Tail Fin Aerodynamics pull request OpenFAST#1874

149b7c8

abhineet-gupta mentioned this pull request Aug 16, 2024

Tail fin aerodynamics: address changes requested in pull request #1874 #2375

Merged

andrew-platt added a commit that referenced this pull request Aug 26, 2024

Merge pull request #2375 from abhineet-gupta/f/tail_fin_unsteadymodel

4daa9df

Tail fin aerodynamics: address changes requested in pull request #1874

andrew-platt mentioned this pull request Dec 24, 2024

Release 4.0.0 #2586

Merged

38 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add unsteady aerodynamic model for turbine tail fin #1874

Add unsteady aerodynamic model for turbine tail fin #1874

abhineet-gupta commented Nov 15, 2023 •

edited

Loading

ebranlard left a comment •

edited

Loading

abhineet-gupta commented Feb 15, 2024

andrew-platt commented Mar 8, 2024

jjonkman left a comment

jjonkman Apr 2, 2024

jjonkman Apr 2, 2024

jjonkman Apr 2, 2024

jjonkman Apr 2, 2024

jjonkman Apr 2, 2024

jjonkman Apr 2, 2024

jjonkman Apr 2, 2024

jjonkman Apr 2, 2024

jjonkman Apr 2, 2024



		where :math:`\sigma_i` are empirical constants characterizing the decay of separation functions, :math:`\gamma_{tf}` is the yaw angle of the tail fin with respect to the free-stream wind (:math:`V_{\text{wind}}`), :math:`\alpha^*_i` are the characteristics angles for separation functions.
		:math:`x_i` takes on a value between 0 and 1, and are used to activate or deactivate a the contribution of potential lift, vortex lift and draft to the normal force on the tail fin.


		.. math:: :label: TFUSBForce

		N = \frac{\rho}{2} A_{tf} \bigg( K_p x_1 V_{\text{rel},x} V_{\text{rel},y} + \Big[x_2 K_v+(1- x_3)C_{Dc} \Big] V_{\text{rel},y}\big\|V_{\text{rel},y}\big\|\bigg)


		The theory will be implemented and documented in a future release.
		x_i = (1+exp{[\sigma_i (\|\gamma_{tf}\|-\alpha^*_i)]})^{-1}

Add unsteady aerodynamic model for turbine tail fin #1874

Add unsteady aerodynamic model for turbine tail fin #1874

Conversation

abhineet-gupta commented Nov 15, 2023 • edited Loading

ebranlard left a comment • edited Loading

Choose a reason for hiding this comment

abhineet-gupta commented Feb 15, 2024

andrew-platt commented Mar 8, 2024

jjonkman left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

abhineet-gupta commented Nov 15, 2023 •

edited

Loading

ebranlard left a comment •

edited

Loading