1 README for wind turbine drag parameterization schemes, added 20101220. JM
2 The code was modified to introduce the power curve and the thrust
3 coefficient using tables. The parameterization is activated in the
4 physics part of the namelist.
5 NOTE: For V3.6 (April 2014) the namelist options and idealized set-up
6 have been changed. Further description below and in README.namelist.
7 The scheme is now activated with windfarm_opt = 1 in the physics namelist.
10 *Specific note for the Fitch scheme.*
12 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
14 !! The following paper should be cited whenever presenting results using this scheme
15 !! (using either the original version or any modified versions of the scheme):
16 !! Fitch, A. C. et al. 2012: Local and Mesoscale Impacts of Wind Farms as Parameterized in a
17 !! Mesoscale NWP Model. Monthly Weather Review, doi:http://dx.doi.org/10.1175/MWR-D-11-00352.1
19 !! Anna C. Fitch, National Center for Atmospheric Research (formerly University of Bergen)
20 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
22 The Fitch scheme is based on Blahak et al. (2010) of Wetter Jetzt GbR.
23 It differs in that the kinetic energy (KE) extracted is based on the
24 thrust coefficient of the turbine (representing the total fraction of KE extracted
25 from the atmosphere due to the turbine), rather than the power coefficient plus an
26 estimated loss factor. In addition the TKE generated is a function of wind speed.
27 The thrust coefficients and the power curve should be obtained from the turbine manufacturer
28 for the turbines of interest and incorporated into a table as described below.
29 If you do not have access to these data an idealized wind turbine is provided in this
30 release (./run/wind-turbine-1.tbl). It is strongly recomended to use the data from the manufacturer.
31 The parameterization works with the MYNN PBL and modifies the QKE field representing 2xTKE
32 to include the TKE produced by wind turbines. QKE can be advected as a part
33 of the scalar 4D tracer array in WRF using the bl_mynn_tkeadvect switch for each domain (default =.false.)
35 It is not recommended to run the parameterization at resolutions higher than 5 rotor diameters -
36 at this resolution rotation of the rotor blades should be included (not represented in the current
37 model). See 1st reference below and Wu and Porte-Agel (2011) and Porte-Agel et al. (2011).
39 The specific implementation comes from Anna Fitch, Alok Gupta, and Idar Barstad at Uni Bergen,
40 Norway. It was added to this release of WRF by Jimy Dudhia, Joe Olson
41 (NOAA), Julie Lundquist (U. Colorado/NREL), and John Michalakes (NREL).
42 Further development comes from Pedro A. Jimenez (CIEMAT/NCAR) who introduced the capability
43 of introducing the manufacturer information via user specified tables as well as other functionalities.
46 Fitch, A. C. et al. 2012: Local and Mesoscale Impacts of Wind Farms as Parameterized in a
47 Mesoscale NWP Model. Monthly Weather Review, doi:http://dx.doi.org/10.1175/MWR-D-11-00352.1
48 Fitch, A. C. et al. 2013: Mesoscale Influences of Wind Farms Throughout a Diurnal Cycle.
49 Monthly Weather Review, doi:http://dx.doi.org/10.1175/MWR-D-12-00185.1
50 Fitch, A. C. et al. 2013: Parameterization of Wind Farms in Climate Models.
51 Journal of Climate, doi:http://dx.doi.org/10.1175/JCLI-D-12-00376.1
52 Jimenez, P.A., J. Navarro, A.M. Palomares and J. Dudhia: Mesoscale modeling of offshore
53 wind turbines wakes at the wind farm resolving scale: a composite-based analysis with
54 the WRF model over Horns Rev. Wind Energy, (In Press.). DOI: 10.1002/we.1708
56 *Compiling the code.* no different from normal
58 *Running the code* To activate the parameterization the user needs
59 to set the domain dependent variable windfarm_opt to 1 in the physics part of the
60 namelist. WRF is expected to find a file called "windturbines.txt". Each line of
61 the file specifies one turbine. The entries of a line are separated by spaces.
62 The entries are listed in order on the line and specify the following,
63 by position on the line:
65 1. True latitude of the wind turbine in degrees [real]
66 2. True longitude of the wind turbine in degrees [real]
67 3. Turbine type [integer]
72 30.000000 -77.000000 1
74 The location of each turbine is specified using the lat and lon elements
75 for the turbine's entry in the windturbines.txt file. That is, the i, j index
76 in the grid is computed from the true latitude (entry 1) and the true
77 longitude (entry 2). The type of turbine (entry 3) points to the file that
78 contains the turbine specifications. If the turbine type is set to 1, WRF is expecting
79 to find the file wind-turbine-1.tbl. The first line of the table is an integer,
80 N, indicating the number of pairs entries for the power curve and the
81 thrust coefficient; the second line contains 4 real values specifying the
82 characteristics of the turbine:
84 1.- Height in meters of the turbine hub [real]
85 2.- Diameter in meters of the rotor [real]
86 3.- Standing thrust coefficient [real]
87 4.- Nominal power of turbine (MW) [real]
89 and the following N lines contain 3 real values with the
90 wind speed, thrust coefficient and power production (kW).
103 *Pseudo-real configuration.* This mode is only for testing the real-world
104 specification mechanism. This option allows to define the position of the
105 turbines using the i and j coordinates on the grid instead of the latitude
106 and longitude. To activate this option the user needs to set the variable
107 windfarm_ij to 1 in the physics part of the namelist. WRF is expecting to
108 find a file "windturbines-ij.txt". Each line of the file is associated with
109 a turbine and each line has three columns corresponding to the I (entry 1), J
110 (entry 2) and kind of turbine (entry 3).