PyWAsP: Working with a WAsP workspace (WWH file)#
This tutorial covers how to read a WAsP workspace hierarchy (.wwh) file with windkit, and how to use its contents in pywasp calculations. We will use a workspace of the Serra Santa Luzia site in Portugal, the same site as in the first and fourth tutorials, saved by WAsP 12.10 with default settings. It holds everything needed for a classic WAsP wind farm calculation:
a met station with its observed wind climate,
a vector map with elevation contours and roughness-change lines,
a wind turbine generator and a group of 15 turbine sites,
a reference site,
and also every result WAsP calculated from them: the generalised wind climate, the predicted wind climates and site effects at each site, and the annual energy production (AEP) of the turbines.
We will first import the inputs and the WAsP results as xarray datasets, then repeat the calculation with pywasp and compare the two. Finally, we will use the same inputs to calculate a resource grid over the wind farm.
After importing the necessary libraries, let’s read the wwh file and list its contents:
[1]:
import warnings
warnings.filterwarnings('ignore') # We will ignore warnings to avoid cluttering the notebook
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import windkit as wk
import pywasp as pw
[2]:
wwh = wk.Workspace.read_wwh('./data/import/ssl_default_12_10_0_30.wwh')
wwh
[2]:
WAsP workspace ssl_default_12_10_0_30.wwh (format version 4.01.0006, CRS EPSG:32629)
ID Object Name
0 WAsP workspace ssl_default_12_10_0_30
1 WAsP project Project 1
2 Turbine site group Turbine cluster
3 Turbine site T1
4 Turbine site T2
5 Turbine site T3
6 Turbine site T4
7 Turbine site T5
8 Turbine site T6
9 Turbine site T7
10 Turbine site T8
11 Turbine site T9
12 Turbine site T10
13 Turbine site T11
14 Turbine site T12
15 Turbine site T13
16 Turbine site T14
17 Turbine site T15
18 Reference site Reference site for the turbine cluster
19 Terrain analysis (IBZ) Terrain analysis (IBZ) 1
20 Vector map Serra Santa Luzia
21 Vector map data layer Serra Santa Luzia
22 Vector map data layer Serra Santa Luzia
23 Generalised wind climate GWC
24 Met. station SerraSantaluzia
25 Observed wind climate SerraSantaLuzia
26 Wind turbine generator Bonus 1 MW
27 Reference site Reference site
The listing shows the WAsP hierarchy as a tree: every object has an ID, its type (the Object column) and the name given to it in WAsP. The first line also tells us the coordinate reference system (CRS) of the workspace, which windkit reads from the header of the vector map. Every dataset we import below is positioned in this CRS, so we don’t need to look up EPSG codes ourselves:
[3]:
print("CRS of the workspace:", wwh.crs)
print("Met. station:", wwh.mast_coords)
CRS of the workspace: 32629
Met. station: {'height': 25.3, 'west_east': 514679.7, 'south_north': 4620970.0, 'description': 'SerraSantaluzia', 'crs': 32629}
Finding objects in the workspace#
Objects are imported by their ID. Instead of reading them off the listing, we can look them up by their type with wwh.get_ids(), which returns a list with the IDs of all objects of that type. This workspace has two reference sites: one in the project and one in the turbine site group. We pick the project’s by its parent in wwh.catalogue:
[4]:
mast_id = wwh.get_ids('Met. station')[0]
group_id = wwh.get_ids('Turbine site group')[0]
turbine_ids = wwh.get_ids('Turbine site')
(ref_id,) = [ID for ID in wwh.get_ids('Reference site') if wwh.catalogue[ID]['ParentID'] != group_id]
print(f"Met. station: {mast_id}, reference site: {ref_id}, turbine site group: {group_id}, turbine sites: {turbine_ids}")
Met. station: 24, reference site: 27, turbine site group: 2, turbine sites: [3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17]
ID note: the
IDis optional in all thewwh.get_methods we use below. Without it, the method imports the only object of its kind in the workspace, and raises an error listing the candidates if there is more than one. This workspace has a single observed wind climate, map, turbine generator and turbine site group, so most of the calls below could be written without anID. We pass it anyway where it makes the example clearer.
Importing the inputs#
The observed wind climate#
We import the Observed wind climate with wwh.get_owc(). It returns a binned wind climate, positioned at the met station that holds it in the workspace tree. As an exercise, we’ll also export it as a NetCDF file, a format commonly used in meteorological applications and large wind atlases:
[5]:
bwc = wwh.get_owc()
print(bwc)
bwc.to_netcdf('./data/export/bwc.nc')
<xarray.Dataset> Size: 4kB
Dimensions: (point: 1, sector: 12, wsbin: 32)
Coordinates:
height (point) float64 8B 25.3
south_north (point) float64 8B 4.621e+06
west_east (point) float64 8B 5.147e+05
* sector (sector) float64 96B 0.0 30.0 60.0 90.0 ... 270.0 300.0 330.0
sector_ceil (sector) float64 96B 15.0 45.0 75.0 ... 285.0 315.0 345.0
sector_floor (sector) float64 96B 345.0 15.0 45.0 ... 255.0 285.0 315.0
* wsbin (wsbin) float64 256B 0.5 1.5 2.5 3.5 ... 28.5 29.5 30.5 31.5
wsceil (wsbin) float64 256B 1.0 2.0 3.0 4.0 ... 29.0 30.0 31.0 32.0
wsfloor (wsbin) float64 256B 0.0 1.0 2.0 3.0 ... 28.0 29.0 30.0 31.0
crs int8 1B 0
Dimensions without coordinates: point
Data variables:
wdfreq (sector, point) float64 96B 0.05314 0.03321 ... 0.1148 0.0707
wsfreq (wsbin, sector, point) float64 3kB 0.02601 0.04219 ... 0.0 0.0
Attributes:
Conventions: CF-1.8
history: 2026-10-01T12:51:21+00:00:\twindkit==0.1.0.dev1+g565f93...
description: SerraSantaluzia
Package name: windkit
Package version: 0.1.0.dev1+g565f937b5
Creation date: 2026-10-01T12:51:21+00:00
Object type: Binned Wind Climate
author: Default User
author_email: default_email@example.com
institution: Default Institution
The vector map#
The workspace has one Vector map object with two Vector map data layer children: one with the elevation contours and one with the roughness-change lines. In WAsP both are used through the single map object. In pywasp we read them as two separate maps, with wwh.get_elevation_map() and wwh.get_roughness_map(), and bundle them together in a pw.wasp.TopographyMap, which is needed for the flow-perturbation calculations.
wwh.get_roughness_map() can return the roughness map either as polygons (the default) or as roughness-change lines, like WAsP uses. To use the default Polygons is recommended.
[6]:
elev_map = wwh.get_elevation_map()
rough_map = wwh.get_roughness_map()
print(elev_map.head())
print(rough_map.head())
topo_map = pw.wasp.TopographyMap(elev_map, rough_map)
elev geometry
0 0.1 LINESTRING (512460.3 4637000, 512464.4 4636988...
1 0.1 LINESTRING (515280.7 4608000, 515277.1 4608009...
2 0.1 LINESTRING (517893.5 4616000, 517895.9 4615991...
3 0.1 LINESTRING (516457.4 4616000, 516454.6 4615989...
4 0.1 LINESTRING (515413.3 4616000, 515412.8 4615993...
id geometry z0 d desc
0 0.0 POLYGON ((520129.5 4608000, 520157.1 4608018.5... 0.03 0.0
1 0.0 POLYGON ((520157.1 4608018.5, 520129.5 4608000... 0.03 0.0
2 0.0 POLYGON ((522602.4 4608000, 522613.5 4608058, ... 0.03 0.0
3 1.0 POLYGON ((522613.5 4608058, 522602.4 4608000, ... 0.60 0.0
4 1.0 POLYGON ((518024.6 4608000, 518062.3 4608074.5... 0.60 0.0
The wind turbine generator and the turbine sites#
The Wind turbine generator is imported with wwh.get_wtg(), which gives the same dataset as wk.read_wtg does for a .wtg file. The turbine positions and hub heights come from the Turbine site group, with wwh.get_turbines():
[7]:
wtg = wwh.get_wtg()
print(wtg.name.item())
turbines = wwh.get_turbines(group_id)
print(turbines)
Bonus 1 MW
<xarray.Dataset> Size: 481B
Dimensions: (point: 15)
Coordinates:
height (point) float64 120B 50.0 50.0 50.0 50.0 ... 50.0 50.0 50.0
south_north (point) float64 120B 4.622e+06 4.622e+06 ... 4.624e+06
west_east (point) float64 120B 5.139e+05 5.142e+05 ... 5.163e+05
crs int8 1B 0
Dimensions without coordinates: point
Data variables:
output (point) float64 120B 0.0 0.0 0.0 0.0 0.0 ... 0.0 0.0 0.0 0.0
Attributes:
Conventions: CF-1.8
history: 2026-10-01T12:51:23+00:00:\twindkit==0.1.0.dev1+g565f937b5\...
The mesoscale climate and the air density#
To generalise a wind climate, WAsP needs the atmospheric stability and baroclinicity of the site, which it looks up in its global mesoscale climate database. Since WAsP 12.8, the workspace stores the values it used. wwh.get_mesoclimate() returns them with the pywasp variable names, so we can pass them straight to pw.wasp.generalize:
[8]:
mesoclimate = wwh.get_mesoclimate()
print(mesoclimate)
<xarray.Dataset> Size: 1kB
Dimensions: (point: 1, sector: 12)
Coordinates:
height (point) float64 8B 25.3
south_north (point) float64 8B 4.621e+06
west_east (point) float64 8B 5.147e+05
* sector (sector) float64 96B 0.0 30.0 60.0 ... 300.0 330.0
sector_ceil (sector) float64 96B 15.0 45.0 75.0 ... 315.0 345.0
sector_floor (sector) float64 96B 345.0 15.0 45.0 ... 285.0 315.0
crs int8 1B 0
Dimensions without coordinates: point
Data variables:
mean_temp_scale_land (sector, point) float64 96B 0.0002808 ... 0.03341
rms_temp_scale_land (sector, point) float64 96B 0.04385 0.0403 ... 0.04724
mean_temp_scale_sea (sector, point) float64 96B -0.01386 ... -0.001693
rms_temp_scale_sea (sector, point) float64 96B 0.04861 ... 0.03138
mean_pblh_scale_land (sector, point) float64 96B 900.7 639.9 ... 1.037e+03
mean_pblh_scale_sea (sector, point) float64 96B 1.603e+03 ... 1.802e+03
mean_dgdz (sector, point) float64 96B 0.007907 ... 0.00851
mean_dgdz_dir (sector, point) float64 96B -36.37 15.28 ... -76.17
Attributes:
Conventions: CF-1.8
history: 2026-10-01T12:51:23+00:00:\twindkit==0.1.0.dev1+g565f93...
Package name: windkit
Package version: 0.1.0.dev1+g565f937b5
Creation date: 2026-10-01T12:51:23+00:00
author: Default User
author_email: default_email@example.com
institution: Default Institution
Likewise, wwh.get_air_density_reference() gives the pressure, temperature and humidity from which WAsP derived the air density at the site:
[9]:
print(wwh.get_air_density_reference())
<xarray.Dataset> Size: 73B
Dimensions: (point: 1)
Coordinates:
height (point) float64 8B 25.3
south_north (point) float64 8B 4.621e+06
west_east (point) float64 8B 5.147e+05
crs int8 1B 0
Dimensions without coordinates: point
Data variables:
source_elevation (point) float64 8B 127.5
source_elevation_temperature (point) float64 8B 129.5
surface_pressure (point) float64 8B 1.003e+05
temperature_celcius (point) float64 8B 14.55
relative_humidity (point) float64 8B 77.42
lapse_rate (point) float64 8B -0.005174
Attributes:
Conventions: CF-1.8
history: 2026-10-01T12:51:23+00:00:\twindkit==0.1.0.dev1+g565f93...
Package name: windkit
Package version: 0.1.0.dev1+g565f937b5
Creation date: 2026-10-01T12:51:23+00:00
author: Default User
author_email: default_email@example.com
institution: Default Institution
The WAsP parameters#
Finally, wwh.get_parameters() returns the WAsP parameters of every object as a pandas DataFrame, and whether each one still has WAsP’s default value. This workspace was made with default settings, so no parameter differs from its default:
[10]:
parameters = wwh.get_parameters()
print(parameters.head())
print("Parameters that differ from WAsP's default:", len(parameters.query('not `Is default`')))
Object ID Object description Source Parameter Value Is default
0 1 WAsP project settings DSECTORS 12.0 True
1 1 WAsP project settings NUMSTDH 5.0 True
2 1 WAsP project settings NUMSTDR 5.0 True
3 1 WAsP project settings STDH1 10.0 True
4 1 WAsP project settings STDH2 25.0 True
Parameters that differ from WAsP's default: 0
Importing the WAsP results#
Besides the inputs, the workspace stores what WAsP calculated from them. These can be imported as well, so they can be analysed in Python or compared with other models.
The generalised wind climate#
We import the Generalised wind climate with wwh.get_gwc(), and export it to NetCDF too:
[11]:
gwc_wasp = wwh.get_gwc()
print(gwc_wasp)
gwc_wasp.to_netcdf('./data/export/gwc.nc')
<xarray.Dataset> Size: 8kB
Dimensions: (point: 1, gen_height: 5, gen_roughness: 5, sector: 12)
Coordinates:
height (point) float64 8B 25.3
south_north (point) float64 8B 4.621e+06
west_east (point) float64 8B 5.147e+05
* gen_height (gen_height) float64 40B 10.0 25.0 50.0 100.0 200.0
* gen_roughness (gen_roughness) float64 40B 0.0 0.03 0.1 0.4 1.5
* sector (sector) float64 96B 0.0 30.0 60.0 90.0 ... 270.0 300.0 330.0
sector_ceil (sector) float64 96B 15.0 45.0 75.0 ... 285.0 315.0 345.0
sector_floor (sector) float64 96B 345.0 15.0 45.0 ... 255.0 285.0 315.0
crs int8 1B 0
Dimensions without coordinates: point
Data variables:
A (gen_height, gen_roughness, sector, point) float64 2kB 5.6...
k (gen_height, gen_roughness, sector, point) float64 2kB 1.7...
wdfreq (gen_height, gen_roughness, sector, point) float64 2kB 0.0...
Attributes:
description: SerraSantaluzia
Conventions: CF-1.8
Package name: windkit
Package version: 0.1.0.dev1+g565f937b5
Creation date: 2026-10-01T12:51:23+00:00
Object type: Weibull Wind Climate
author: Default User
author_email: default_email@example.com
institution: Default Institution
history: 2026-10-01T12:51:23+00:00:\twindkit==0.1.0.dev1+g565f93...
Predicted wind climates and site effects#
wwh.get_wwc() returns the predicted wind climate (the sector-wise Weibull A and k, the sector frequencies and the air density) and wwh.get_site_effects() the site effects (speed-ups, turnings, RIX, mesoscale roughness, …). Both accept the ID of a single site, or of a turbine site group to get one point per turbine site:
[12]:
wwc_ref_wasp = wwh.get_wwc(ref_id)
site_effects_ref_wasp = wwh.get_site_effects(ref_id)
print(wwc_ref_wasp)
<xarray.Dataset> Size: 617B
Dimensions: (point: 1, sector: 12)
Coordinates:
height (point) float64 8B 100.0
south_north (point) float64 8B 4.621e+06
west_east (point) float64 8B 5.137e+05
name (point) object 8B 'Reference site'
* sector (sector) float64 96B 0.0 30.0 60.0 90.0 ... 270.0 300.0 330.0
sector_ceil (sector) float64 96B 15.0 45.0 75.0 ... 285.0 315.0 345.0
sector_floor (sector) float64 96B 345.0 15.0 45.0 ... 255.0 285.0 315.0
crs int8 1B 0
Dimensions without coordinates: point
Data variables:
A (sector, point) float64 96B 6.809 6.414 5.823 ... 7.166 6.797
k (sector, point) float64 96B 2.084 2.346 2.568 ... 2.115 2.002
wdfreq (sector, point) float64 96B 0.06475 0.04059 ... 0.08201
air_density (point) float64 8B 1.178
Attributes:
Conventions: CF-1.8
history: 2026-10-01T12:51:23+00:00:\twindkit==0.1.0.dev1+g565f93...
Package name: windkit
Package version: 0.1.0.dev1+g565f937b5
Creation date: 2026-10-01T12:51:23+00:00
Object type: Weibull Wind Climate
author: Default User
author_email: default_email@example.com
institution: Default Institution
description: WAsP predicted wind climate at reference site 'Referenc...
Several objects in this workspace hold a predicted wind climate (the met station, the reference site and each turbine site), so here the ID is not optional. Calling wwh.get_wwc() without it raises an error that lists the candidates:
[13]:
try:
wwh.get_wwc()
except ValueError as error:
print(error)
The workspace holds 19 'Met. station', 'Reference site', 'Turbine site' or 'Turbine site group' objects, so ID must be given. Candidates:
ID Object Name
2 Turbine site group Turbine cluster
3 Turbine site T1
4 Turbine site T2
5 Turbine site T3
6 Turbine site T4
7 Turbine site T5
8 Turbine site T6
9 Turbine site T7
10 Turbine site T8
11 Turbine site T9
12 Turbine site T10
13 Turbine site T11
14 Turbine site T12
15 Turbine site T13
16 Turbine site T14
17 Turbine site T15
18 Reference site Reference site for the turbine cluster
24 Met. station SerraSantaluzia
27 Reference site Reference site
Annual energy production#
wwh.get_aep() returns the gross and the potential AEP of the turbine sites in GWh/y. WAsP calls the AEP after wake losses its “net” AEP; in pywasp this is the potential_aep, since further losses are still to be subtracted (see tutorial 6):
[14]:
aep_wasp = wwh.get_aep(group_id)
print(aep_wasp)
print(f"Wind farm gross AEP: {aep_wasp.gross_aep.sum().item():.2f} GWh/y, potential AEP: {aep_wasp.potential_aep.sum().item():.2f} GWh/y")
<xarray.Dataset> Size: 4kB
Dimensions: (point: 15, sector: 12)
Coordinates:
height (point) float64 120B 50.0 50.0 50.0 ... 50.0 50.0 50.0
south_north (point) float64 120B 4.622e+06 4.622e+06 ... 4.624e+06
west_east (point) float64 120B 5.139e+05 5.142e+05 ... 5.163e+05
name (point) object 120B 'T1' 'T2' 'T3' ... 'T14' 'T15'
* sector (sector) float64 96B 0.0 30.0 60.0 ... 300.0 330.0
sector_ceil (sector) float64 96B 15.0 45.0 75.0 ... 315.0 345.0
sector_floor (sector) float64 96B 345.0 15.0 45.0 ... 285.0 315.0
crs int8 1B 0
Dimensions without coordinates: point
Data variables:
gross_aep_sector (sector, point) float64 1kB 0.1428 0.1423 ... 0.2183
gross_aep (point) float64 120B 2.796 2.66 2.448 ... 3.063 3.277
potential_aep_sector (sector, point) float64 1kB 0.1428 0.1423 ... 0.2183
potential_aep (point) float64 120B 2.695 2.568 2.341 ... 2.895 3.157
Attributes:
Conventions: CF-1.8
history: 2026-10-01T12:51:23+00:00:\twindkit==0.1.0.dev1+g565f93...
Package name: windkit
Package version: 0.1.0.dev1+g565f937b5
Creation date: 2026-10-01T12:51:23+00:00
Object type: Anual Energy Production
author: Default User
author_email: default_email@example.com
institution: Default Institution
description: WAsP annual energy production at turbine site group 'Tu...
Wind farm gross AEP: 41.64 GWh/y, potential AEP: 40.28 GWh/y
warnings note: we switched off warnings at the start of this notebook, but the result getters warn when WAsP marked results as not up to date (e.g., the inputs were changed after the calculation). Recalculate in WAsP and save the workspace before reading its results. All results in this workspace are up to date.
Anything else: the low-level member data#
The wwh.get_ methods cover what pywasp needs as datasets. Everything else stored for an object can be read from wwh.get_member(), which returns a plain dictionary. For example, the wake model settings of the turbine site group:
[15]:
wake_modelling = wwh.get_member(group_id)['wake_modelling']
print(wake_modelling)
{'is_active': True, 'model': 'Park2', 'n_sub_sectors': 5, 'speed_resolution': np.float64(1.0), 'default_decay_coefficient': np.float64(0.09), 'decay_coefficient': array([0.09, 0.09, 0.09, 0.09, 0.09, 0.09, 0.09, 0.09, 0.09, 0.09, 0.09,
0.09])}
Repeating the WAsP calculation with pywasp#
We now have all the inputs WAsP used, so let’s repeat its calculation with pywasp and see how the results compare.
The workspace was saved by WAsP 12.10. pywasp’s default configuration, pw.wasp.Config(), uses its WAsP_12.10 parameter set, which reproduces WAsP 12.10. pw.wasp.generalize also defaults to WAsP’s standard generalisation heights, the same heights as the generalised wind climate in the workspace:
[16]:
conf = pw.wasp.Config()
print("Generalisation heights:", gwc_wasp.gen_height.values)
Generalisation heights: [ 10. 25. 50. 100. 200.]
The wind directions of the observed wind climate are measured relative to the grid north of the map, as in WAsP, so we tell pywasp by setting the wind_dir_crs attribute to the CRS of the map. We pass the mesoscale climate from the workspace, so pywasp uses exactly the same stability and baroclinicity as WAsP:
[17]:
bwc.attrs['wind_dir_crs'] = wk.spatial.get_crs(bwc).to_wkt()
gwc = pw.wasp.generalize(bwc, topo_map, conf=conf, mesoclimate=mesoclimate)
At line 555 of file ../../modules/waspcore/Rvea0288/Internal/generalization.f90
Fortran runtime warning: An array temporary was created for argument 'gd' of procedure 'gmom_ntb'
At line 555 of file ../../modules/waspcore/Rvea0288/Internal/generalization.f90
Fortran runtime warning: An array temporary was created for argument 'gf' of procedure 'gmom_ntb'
At line 586 of file ../../modules/waspcore/Rvea0288/Internal/generalization.f90
Fortran runtime warning: An array temporary was created for argument 'gd' of procedure 'gmom_ntb'
At line 586 of file ../../modules/waspcore/Rvea0288/Internal/generalization.f90
Fortran runtime warning: An array temporary was created for argument 'gf' of procedure 'gmom_ntb'
Next we downscale the generalised wind climate to the turbine sites, and calculate their gross AEP and, with WAsP’s wake model settings, their potential AEP. WAsP’s wake model is the PARK2_onshore wind farm model in pywasp:
[18]:
pwc_turbines = pw.wasp.downscale(gwc, topo_map, turbines, conf=conf, return_site_effects=True)
gross_aep = pw.gross_aep(pwc_turbines, wtg, air_density_correction='infer')
potential_aep = pw.potential_aep(
pwc_turbines,
wtg,
air_density_correction='infer',
wind_farm_model='PARK2_onshore',
n_subsector=wake_modelling['n_sub_sectors'],
ws_stepsize=wake_modelling['speed_resolution'],
)
Let’s put the AEP of each turbine from both models side by side:
[19]:
aep_comparison = pd.DataFrame(
{
'Gross WAsP': aep_wasp.gross_aep.values,
'Gross PyWAsP': gross_aep.gross_aep.values,
'Potential WAsP': aep_wasp.potential_aep.values,
'Potential PyWAsP': potential_aep.potential_aep.values,
},
index=aep_wasp.name.values,
)
aep_comparison.loc['Wind farm'] = aep_comparison.sum()
aep_comparison.round(4)
[19]:
| Gross WAsP | Gross PyWAsP | Potential WAsP | Potential PyWAsP | |
|---|---|---|---|---|
| T1 | 2.7958 | 2.7957 | 2.6951 | 2.6950 |
| T2 | 2.6601 | 2.6600 | 2.5682 | 2.5682 |
| T3 | 2.4484 | 2.4483 | 2.3408 | 2.3407 |
| T4 | 2.5325 | 2.5325 | 2.4229 | 2.4229 |
| T5 | 2.6970 | 2.6970 | 2.5877 | 2.5877 |
| T6 | 2.7743 | 2.7743 | 2.7290 | 2.7289 |
| T7 | 2.6528 | 2.6528 | 2.6219 | 2.6218 |
| T8 | 2.9498 | 2.9497 | 2.8987 | 2.8987 |
| T9 | 3.0417 | 3.0417 | 2.9793 | 2.9792 |
| T10 | 2.2840 | 2.2840 | 2.2102 | 2.2102 |
| T11 | 2.6803 | 2.6802 | 2.6201 | 2.6201 |
| T12 | 2.9138 | 2.9138 | 2.8259 | 2.8258 |
| T13 | 2.8710 | 2.8709 | 2.7241 | 2.7241 |
| T14 | 3.0630 | 3.0629 | 2.8948 | 2.8948 |
| T15 | 3.2769 | 3.2768 | 3.1568 | 3.1568 |
| Wind farm | 41.6413 | 41.6407 | 40.2755 | 40.2749 |
The two models agree to within a few thousandths of a percent, both for the individual turbines and for the whole wind farm.
Calculating a resource grid over the wind farm#
With the inputs from the workspace we are not limited to the sites defined in WAsP. Let’s create a uniform grid with a resolution of 100 m around the turbines, at their hub height of 50 m, and downscale the generalised wind climate to it:
[20]:
height = 50
res = 100
output_locs = wk.spatial.create_dataset(
np.arange(512500, 517000 + res, res),
np.arange(4620500, 4625000 + res, res),
np.array([height]),
wwh.crs,
)
pwc_grid = pw.wasp.downscale(gwc, topo_map, output_locs, conf=conf)
print(pwc_grid)
pwc_grid.to_netcdf(f'./data/export/results_{height}_m.nc')
<xarray.Dataset> Size: 340kB
Dimensions: (sector: 12, height: 1, south_north: 46, west_east: 46)
Coordinates:
* sector (sector) float64 96B 0.0 30.0 60.0 90.0 ... 270.0 300.0 330.0
sector_ceil (sector) float64 96B 15.0 45.0 75.0 ... 285.0 315.0 345.0
sector_floor (sector) float64 96B 345.0 15.0 45.0 ... 255.0 285.0 315.0
* height (height) int64 8B 50
* south_north (south_north) int64 368B 4620500 4620600 ... 4624900 4625000
* west_east (west_east) int64 368B 512500 512600 512700 ... 516900 517000
crs int8 1B 0
Data variables:
A (sector, height, south_north, west_east) float32 102kB 4.6...
k (sector, height, south_north, west_east) float32 102kB 2.0...
wdfreq (sector, height, south_north, west_east) float32 102kB 0.0...
site_elev (south_north, west_east) float32 8kB 199.8 194.9 ... 252.1
air_density (height, south_north, west_east) float32 8kB 1.194 ... 1.187
wspd (height, south_north, west_east) float32 8kB 6.238 ... 5.179
power_density (height, south_north, west_east) float32 8kB 262.7 ... 160.1
Attributes:
Conventions: CF-1.8
history: 2026-10-01T12:51:25+00:00:\twindkit==0.1.0.dev1+g565f93...
title: WAsP site effects
Package name: windkit
Package version: 0.1.0.dev1+g565f937b5
Creation date: 2026-10-01T12:51:45+00:00
Object type: Met fields
author: Default User
author_email: default_email@example.com
institution: Default Institution
And plot the mean wind speed and power density, with the turbine sites (black) and the met station (red) on top:
[21]:
fig, axes = plt.subplots(1, 2, figsize=(17, 6))
ax1, ax2 = axes.flat
variables = ['wspd', 'power_density']
for var, ax in zip(variables, axes.flat):
pwc_grid[var].isel(height=0).plot(ax=ax, cmap='coolwarm')
ax.scatter(turbines.west_east, turbines.south_north, c='k', s=15)
ax.scatter(bwc.west_east, bwc.south_north, c='r', marker='^', s=40)
ax.set_aspect('equal')
plt.setp(ax.get_xticklabels(), rotation=30, horizontalalignment='right')
ax1.set_title("Wind Speed")
ax2.set_title("Power Density")
fig.tight_layout()
