ECN: IEA ANNEX-XIV Introduction Database

ECN

INTRODUCTION DATABASE

Conventions and notations

The prescribed definition of conventions and notations are given in Appendix A of J.G. Schepers, et al.: Final report of IEA ANNEX-XVIII: Enhanced Field Rotor Aerodynamics Database , ECN-report: ECN-C-02-016. In particular the definition of azimuth angle and yaw error turned out to be of importance:

  • The azimuth angle has been defined zero when the instrumented blade (blade 1) is pointing vertically up;
  • For a clockwise rotating wind turbine, the yaw error is defined positive when the wind comes from the East, when the turbine is 'looking' to the North;
  • For a counter clockwise rotating turbine, the yaw error is defined positive when the wind comes from theb West, when the turbine is 'looking' to the North.

Format of the files, filetypes and filenames

In order to facilate the selection of signals from a file, it was considered essential that all files were supplied in the same format. 

Time series

Three types of files had to be supplied:

  • Log files. These files contain general information about the measurement procedure. It should also contain:
    • Date and time of measurement
    • Air density, pressure and temperature
    • Information which is needed to read the other files (like the number of pressure taps, the number of anemometers, etc.)
  • Profile files. These files contain the wind speeds, the wind directions, the rotor speed, the pitch angle, the mechanical loads, the angles of attack and the inflow velocities, the profile coefficients (cl and cd) all as function of time and azimuth angle.
  • Pressure files. These files contain the pressure recordings.

The format of the files is specified. The files start with the statistics (mean, max, min, standard deviation) of every signal.
The filenames are structured as follows: '(n)y_ext1_ext2_ext3_n.rad'

  • (n)y refers to (non)-yawed conditions;
  • ext1 refers to the file type:
    • log = log file
    • prof = file with wind conditions and profile data;
    • p1----pN = files with pressures
  • ext2 refers to the type of experiment:
    • rot = rotating conditions
    • non = non-rotating conditions
  • ext3 refers to the name of the participant:
    • c = CRES
    • d = DUT
    • e = ECN
    • i = IC/RAL
    • m = Mie University
    • nu = NREL untwisted blade, phase II
    • nt = NREL twisted blade, phase III
    • ntp = NREL twisted blade, phase IV
    • r = RISØ
  • n is the unique label of the experiment
  • rad is the radial position (Only for DUT and MIE experiments)

For example ny_p03_rot_nt_001 is the first NREL datafile of phase III which contains the pressure data of the 3rd blade section obtained for non yawed conditions. Actually the filenames may be abbreviated. The abbreviations originate from the period that common operating systems could not support very long filenames.
 

Airfoil coefficients measured under rotating conditions

The rotating airfoil coefficients are supplied in the following form:

  • cn as function of angle of attack
  • ct as function of angle of attack
  • cm as function of angle of attack
  • cn as function of ct

For each characteristic and for each instrumented section, the results are stored in a separate file. The first column in the file is the angle of attack (or the ct). The second column is the coefficient. The filenames are structured as follows:
cxan_rot_ext1 (for cn as function of ct: cnctn_rot_ext1:

  • x identifies the load (n for normal force, t for tangential force, m for moment)
  • Only if applicable: n identifies the section number
  • ext1 refers to the name of the participant, i.e. similar to ext3 for the time series files

Profile and pressure coefficients measured under 2D conditions

The 2D sectional characteristics are supplied as function of angle of attack in a number of separate files: 1 file per coefficient, per section. The first column is the angle of attack ad the second column is the coefficient. The 2D pressure coefficients are supplied as function of angle of attack in 1 file. The first column is the angle of attack, the remaining columns give the pressure coefficient for the different pressure taps.
The file names are structured as follows: cxan_2D_ext1

  • x identifies the load (l for lift, d for drag, m for moment and p for pressure coefficient)
  • Only if applicable: n identifies the section number
  • ext1 refers to the name of the participant, i.e. similar to ext3 for the time series files

Occasionally, the Reynolds number and the presence of probes etc. can be found in the file name.

Power curves

The rotating power curves are supplied as function of wind speed. In these files, the first column gives the windspeed, the second column gives the electrical power. Some paricipants supplied a third column with the mechanical power. The name of the file with the power curve is:

  • PV_ext1, with ext1 the name of the participant (see also the filename of the time series).

Content of database (February 2002)

  • In February 2002, measurements on 6 different facilities were stored.
  • The number of time series files is  more than 400 with a length of 22000 s.
  • The length of the time series range between 15s and 10 minutes. Data over a 10 minute period could be supplied by RISØ, because this institute measured the aerodynamic forces only.
  • The other participants also supplied pressure files. Then measurements from a shorter time period had to be supplied in order to limit the amount of data.
  • Rotating as well as non-rotating experiments are stored.
  • Furthermore the measurements are supplied for non-yawed and yawed conditions.
  • NREL has supplied measurements of the Phase II, III and IV experiments.
  • The angle of attack varies between negative values and deep stall. The adjustment of angle of attack is realized in different ways by the participants:
    • By means of a variation in wind speed; This has been realized by ECN, RISØ, IC/RAL and NREL
    • By means of a variation in rotor speed. This has been realized by DUT and ECN
    • By means of a variation in pitch angle: This has been realized by ECN and Mie

In addition to the time series, measurements of airfoil coefficients were stored as well as 2D airfoil coefficients. In order to limit the required disc capacity, the files are compressed by means of the zip utility. It is expected that the abbreviated file names are self-explanatory, however key files which translate the abbreviated names to the prescribed conventions are available too.

Availability

The IEA ANNEX-XIV/XVIII database is stored on CD-ROM and on this WWW-site. which is protected by means of a password. The CD-ROM and the online database are available for outside parties under the condition that they inform the IEA ANNEX-XIV/XVIII participants about experiences gained with the database. This gives the IEA ANNEX-XIV/XVIII participants the possibilty to improve the quality of the database. By filling out the registration form, access to the database can be obtained.

Users of database

In February 2002, a large number of institutes has visited the database,
Some interesting topics which have been studied using aerodynamic field measurements are:

  • Instationary and 3D Effects at pre-stalled and stalled conditions stall, see i.e. [1]. It was already expected from helicopter and propellor experiences that  the airfoil coefficients at rotation would differ from the 2D airfoil coefficients, i.e. at rotation a much higher maximum lift coefficient was expected in particular at the root of the blade. The field measurements led to a definite confirmation of this expectation;
  • Validation of codes (free wake panel methods,  NS codes, bem-like methods), see i.e. [2], [3], [4] and http://www.ae.gatech.edu/~lsankar/NREL/;
  • The variation of the induced velocities at yawed conditions, see i.e. [5];
  • Rotor tower interaction, see i.e. [6].
  • The effect of different pitch angles on the airfoil characteristics. Indications were found that the differences between 2D and rotating airfoil data depend on the pitch angle.
  • On May 1st, 2002 the Dutch national project Annexlyse will commence. In this project ECN and DUT will cooperate in analysing the IEA Annex XVIII data. The aim of this project is to validate aerodynamic and aeroelastic codes on basis of the IEA Annex XVIII data and to extract information on the field of (i.e.) instationary and 3D effects in stall and yawed conditions. Measurements will be selected at conditions which are as stationary as possible. As soon as results are available, they will be reported on the internet.

References

[1] A. Bjorck, Dynamic Stall and Three Dimensional Effect, FFA-TN-1995-31, FFA, 1995
[2] P.K. Chaviaropoulos et al, Viscous and Aeroelastic effects on Wind turbine blades. The Viscel project. In Proceedings of European Wind Energy Conference, EWEC,  pp 347-351, Copenhagen, July, 2001
[3] P. Strzelczyk, On Simple Vortex Theory of Horizontal Axis Wind Turbines,  Rzeszow University of Technology, Faculty of Mechanical Engineering and Aviation, 1998.
[4] L. Bermudez, A. Velasquez and A. Matesanz Numerical Simulation of Unsteady Aerodynamic Effects  in Horizontal Axis Wind Turbines, Solar Energy, Volume 68, 2000
[5] H.A. Madsen , Yaw simulation using a 3D actuator disc model,  Proceedings, IEA aerodynamics Symposium, Stockholm, November 1999
[6] J.M.R. Graham and C.J. Brown ROTOW-Investigation of the aerodynamic interaction between wind turbine rotor blades and the tower and its impact on wind turbine design, Final Publishable Report, Department of Aeronautics, Imperial College of Science and Medicine, London, UK, September 2000.

ECN, P.O. Box 1, 1755 ZG Petten, tel +31 224 56 4949  |  Disclaimer  |  Privacy Statement