
# Dataset for Article

* Title: Potential of dynamic wind farm control by axial induction in the case of wind gusts

* Authors:
- Florian Buergel
  . TU Braunschweig, Institute for Mathematical Optimization, Germany
  . f.buergel@tu-braunschweig.de. ORCID iD: https://orcid.org/0000-0001-8673-5849.
- Robert Scholz
  . University of Heidelberg, Interdisciplinary Center for Scientific Computing, Germany
  . robert.scholz@iwr.uni-heidelberg.de.
- Christian Kirches
  . TU Braunschweig, Institute for Mathematical Optimization, Germany
  . c.kirches@tu-braunschweig.de. ORCID iD: https://orcid.org/0000-0002-3441-8822.
- Sebastian Stiller
  . TU Braunschweig, Institute for Mathematical Optimization, Germany
  . sebastian.stiller@tu-braunschweig.de. ORCID iD: https://orcid.org/0000-0002-8792-4390.

* Correspondence: Florian Buergel (f.buergel@tu-braunschweig.de)

* Keywords: Wind farm, dynamic wake, optimization, axial-induction-based control, wind gust.

* Submission of revised manuscript: November 2023

-----

# Sharing Information

* License: Creative Commons Attribution License (CC-BY)

-----

# Methodological Information

* Experimental set-up:
  - two wind turbines of type NREL-5-MW
  - in a straight line with distance of 5D
  - yaw offset of 0 deg
  - wind field: turbulence intensity of 0.01 and wind gust with abrupt change of 1 m/s

* Simulation software: FAST.Farm, which is a part of OpenFAST, see NREL (2023)
  - wind field generator: TurbSim, which is also a part of OpenFAST

* Optimization:
  - control variable: blade pitch angle (and corresponding adaption of generator torque)
  - Structure of the objective function:
    min -<power> + <weight of tower activity> <tower activity> + <weight of pitch activity> <pitch activity>

* Workflow: The optimization procedure is implemented as a module of the self-developed and unpublished Wind Farm Optimization software suite WiFaOpt.

-----

# Data Overview

* T1.out and T2.out are the output from FAST.Farm simulation, namely:
  - Time
  - Wind1VelX
  - Wind1VelY
  - Wind1VelZ
  - OoPDefl1
  - IPDefl1
  - RotSpeed
  - GenSpeed
  - NcIMUTVxs
  - TTDspFA
  - TTDspSS
  - RootMIP1
  - RootMOoP1
  - RootMzb1
  - RotTorq
  - YawBrMyp
  - TwrBsMyt
  - YawBrFxn
  - RtSkew
  - RtVAvgxh
  - BlPitchC1
  - GenPwr
  - GenTq

* We always have two data sets:
  - baseline (i.e., greedy control) in 'bas'
  - and optimization in 'opt'

* We have three scenarios for each case, where scenario 3 includes the wind gust:

  case | scenario | wind speed (in m/s)
  -------------------------------------
    1  |     1    |     6
    1  |     2    |     7
    1  |     3    |  6 to 7
  -------------------------------------
    2  |     1    |    11
    2  |     2    |    11
    2  |     3    | 11 to 12
  -------------------------------------
    3  |     1    |    12
    3  |     2    |    12
    3  |     3    | 12 to 13

* the wind speeds in the filename represent the case, i.e.:
  - Uave_06 is case 1
  - Uave_11 is case 2
  - Uave_12 is case 3
  - i.e., Uave_06_scenario_2 includes a wind speed of 7 m/s and scenario 3 a speed of 6 to 7 m/s

-----

# References

- NREL: OpenFAST. Version 3.5.0, https://github.com/OpenFAST/openfast, 2023.

