Advanced Wind Turbine Simulation Workflow¶

• A more advanced turbine workflow will include many steps (Which have been showcased in previous examples)

Workflow:

1. Weather data extraction from a MERRA dataset (windspeed, pressure, temperature)
2. Spatial adjustment of the windspeeds
3. Vertical projection of the wind speeds
4. Wind speed density correction
5. Power curve convolution
6. Capacity Factor Estimation

In [1]:

import reskit as rk

import numpy as np
import matplotlib.pyplot as plt

import reskit as rk import numpy as np import matplotlib.pyplot as plt

Simulate a Single Location¶

In [2]:

# Set some constants for later

TURBINE_CAPACITY = 4200  # kW
TURBINE_HUB_HEIGHT = 120  # meters
TURBINE_ROTOR_DIAMETER = 136  # meters
TURBINE_LOCATION = (6.0, 50.5)  # (lon, lat)

# Set some constants for later TURBINE_CAPACITY = 4200 # kW TURBINE_HUB_HEIGHT = 120 # meters TURBINE_ROTOR_DIAMETER = 136 # meters TURBINE_LOCATION = (6.0, 50.5) # (lon, lat)

In [3]:

# 1. Create a weather source, load, and extract weather variables
src = rk.weather.MerraSource(rk.TEST_DATA["merra-like"], bounds=[5, 49, 7, 52], verbose=False)

src.sload_elevated_wind_speed()
src.sload_surface_pressure()
src.sload_surface_air_temperature()

raw_windspeeds = src.get("elevated_wind_speed", locations=TURBINE_LOCATION, interpolation="bilinear")
raw_pressure = src.get("surface_pressure", locations=TURBINE_LOCATION, interpolation="bilinear")
raw_temperature = src.get("surface_air_temperature", locations=TURBINE_LOCATION, interpolation="bilinear")

print(raw_windspeeds.head())

# 1. Create a weather source, load, and extract weather variables src = rk.weather.MerraSource(rk.TEST_DATA["merra-like"], bounds=[5, 49, 7, 52], verbose=False) src.sload_elevated_wind_speed() src.sload_surface_pressure() src.sload_surface_air_temperature() raw_windspeeds = src.get("elevated_wind_speed", locations=TURBINE_LOCATION, interpolation="bilinear") raw_pressure = src.get("surface_pressure", locations=TURBINE_LOCATION, interpolation="bilinear") raw_temperature = src.get("surface_air_temperature", locations=TURBINE_LOCATION, interpolation="bilinear") print(raw_windspeeds.head())

2015-01-01 00:30:00+00:00    5.383743
2015-01-01 01:30:00+00:00    5.323299
2015-01-01 02:30:00+00:00    5.232418
2015-01-01 03:30:00+00:00    5.160997
2015-01-01 04:30:00+00:00    5.439261
Name: (6.0, 50.5), dtype: float64

In [4]:

# 2. Vertically project wind speeds to hub height

roughness = rk.wind.roughness_from_clc(clc_path=rk.TEST_DATA["clc-aachen_clipped.tif"], loc=TURBINE_LOCATION)


projected_windspeed = rk.wind.apply_logarithmic_profile_projection(
    measured_wind_speed=raw_windspeeds,
    measured_height=50,  # The MERRA dataset offers windspeeds at 50m
    target_height=TURBINE_HUB_HEIGHT,
    roughness=roughness,
)

print(projected_windspeed.head())

# 2. Vertically project wind speeds to hub height roughness = rk.wind.roughness_from_clc(clc_path=rk.TEST_DATA["clc-aachen_clipped.tif"], loc=TURBINE_LOCATION) projected_windspeed = rk.wind.apply_logarithmic_profile_projection( measured_wind_speed=raw_windspeeds, measured_height=50, # The MERRA dataset offers windspeeds at 50m target_height=TURBINE_HUB_HEIGHT, roughness=roughness, ) print(projected_windspeed.head())

2015-01-01 00:30:00+00:00    6.506035
2015-01-01 01:30:00+00:00    6.432992
2015-01-01 02:30:00+00:00    6.323166
2015-01-01 03:30:00+00:00    6.236856
2015-01-01 04:30:00+00:00    6.573127
Name: (6.0, 50.5), dtype: float64

In [5]:

# 3. Apply density correction

pressure_corrected_windspeeds = rk.wind.apply_air_density_adjustment(
    wind_speed=projected_windspeed,
    pressure=raw_pressure,
    temperature=raw_temperature,
    height=TURBINE_HUB_HEIGHT,
)

pressure_corrected_windspeeds.head()

# 3. Apply density correction pressure_corrected_windspeeds = rk.wind.apply_air_density_adjustment( wind_speed=projected_windspeed, pressure=raw_pressure, temperature=raw_temperature, height=TURBINE_HUB_HEIGHT, ) pressure_corrected_windspeeds.head()

Out[5]:

2015-01-01 00:30:00+00:00    6.560702
2015-01-01 01:30:00+00:00    6.485689
2015-01-01 02:30:00+00:00    6.374120
2015-01-01 03:30:00+00:00    6.287352
2015-01-01 04:30:00+00:00    6.626462
Name: (6.0, 50.5), dtype: float64

In [6]:

# 4. Power curve estimation and convolution

power_curve = rk.wind.PowerCurve.from_capacity_and_rotor_diam(
    capacity=TURBINE_CAPACITY, rotor_diam=TURBINE_ROTOR_DIAMETER
)

convoluted_power_curve = power_curve.convolute_by_gaussian(scaling=0.06, base=0.1)

# 4. Power curve estimation and convolution power_curve = rk.wind.PowerCurve.from_capacity_and_rotor_diam( capacity=TURBINE_CAPACITY, rotor_diam=TURBINE_ROTOR_DIAMETER ) convoluted_power_curve = power_curve.convolute_by_gaussian(scaling=0.06, base=0.1)

In [7]:

# 5. Capacity factor estimation
capacity_factors = convoluted_power_curve.simulate(wind_speed=pressure_corrected_windspeeds)

capacity_factors

# 5. Capacity factor estimation capacity_factors = convoluted_power_curve.simulate(wind_speed=pressure_corrected_windspeeds) capacity_factors

Out[7]:

array([0.27273363, 0.26315142, 0.24928885, 0.23878838, 0.2813027 ,
       0.38840875, 0.46433861, 0.50707569, 0.55526571, 0.49867274,
       0.5716173 , 0.71244977, 0.70559868, 0.61191423, 0.70556453,
       0.91621572, 0.98436881, 0.98905482, 0.99521448, 0.99879609,
       0.99956091, 0.99987679, 0.99996843, 0.99999171, 0.99999641,
       0.99999844, 0.99999941, 0.99999979, 0.99999986, 0.99999954,
       0.99999577, 0.99996633, 0.99986441, 0.99835949, 0.99779936,
       0.98064443, 0.94692822, 0.97454365, 0.97228814, 0.97970383,
       0.97176293, 0.93889837, 0.87918241, 0.82917666, 0.78503089,
       0.73943709, 0.6327937 , 0.60788396, 0.56244002, 0.49216547,
       0.48407326, 0.55137321, 0.66608584, 0.71491774, 0.71633097,
       0.77210396, 0.8345534 , 0.76770054, 0.56405061, 0.42807357,
       0.41323894, 0.41240108, 0.32953804, 0.23966265, 0.13567885,
       0.12706234, 0.32103426, 0.49467238, 0.26512148, 0.20028125,
       0.37167453])

In [8]:

plt.plot(capacity_factors)
plt.xlabel("Time")
plt.ylabel("Capacity Factor")
plt.grid()
plt.show()

plt.plot(capacity_factors) plt.xlabel("Time") plt.ylabel("Capacity Factor") plt.grid() plt.show()

Simulate multiple locations at once (recommended)¶

In [9]:

TURBINE_CAPACITY = 4200  # kW
TURBINE_HUB_HEIGHT = 120  # meters
TURBINE_ROTOR_DIAMETER = 136  # meters
TURBINE_LOCATION = np.array([(6.25, 51.0), (6.50, 51.0), (6.25, 50.75)])  # (lon,lat)


# 1
raw_windspeeds = src.get("elevated_wind_speed", locations=TURBINE_LOCATION, interpolation="bilinear")
raw_pressure = src.get("surface_pressure", locations=TURBINE_LOCATION, interpolation="bilinear")
raw_temperature = src.get("surface_air_temperature", locations=TURBINE_LOCATION, interpolation="bilinear")

# 2
roughness = rk.wind.roughness_from_clc(clc_path=rk.TEST_DATA["clc-aachen_clipped.tif"], loc=TURBINE_LOCATION)


projected_windspeed = rk.wind.apply_logarithmic_profile_projection(
    measured_wind_speed=raw_windspeeds,
    measured_height=50,  # The MERRA dataset offers windspeeds at 50m
    target_height=TURBINE_HUB_HEIGHT,
    roughness=roughness,
)

# 3
pressure_corrected_windspeeds = rk.wind.apply_air_density_adjustment(
    wind_speed=projected_windspeed,
    pressure=raw_pressure,
    temperature=raw_temperature,
    height=TURBINE_HUB_HEIGHT,
)

convoluted_power_curve = power_curve.convolute_by_gaussian(scaling=0.06, base=0.1)

# 4
power_curve = rk.wind.PowerCurve.from_capacity_and_rotor_diam(
    capacity=TURBINE_CAPACITY, rotor_diam=TURBINE_ROTOR_DIAMETER
)

# 5
capacity_factors = convoluted_power_curve.simulate(wind_speed=pressure_corrected_windspeeds)


# Print result
capacity_factors

TURBINE_CAPACITY = 4200 # kW TURBINE_HUB_HEIGHT = 120 # meters TURBINE_ROTOR_DIAMETER = 136 # meters TURBINE_LOCATION = np.array([(6.25, 51.0), (6.50, 51.0), (6.25, 50.75)]) # (lon,lat) # 1 raw_windspeeds = src.get("elevated_wind_speed", locations=TURBINE_LOCATION, interpolation="bilinear") raw_pressure = src.get("surface_pressure", locations=TURBINE_LOCATION, interpolation="bilinear") raw_temperature = src.get("surface_air_temperature", locations=TURBINE_LOCATION, interpolation="bilinear") # 2 roughness = rk.wind.roughness_from_clc(clc_path=rk.TEST_DATA["clc-aachen_clipped.tif"], loc=TURBINE_LOCATION) projected_windspeed = rk.wind.apply_logarithmic_profile_projection( measured_wind_speed=raw_windspeeds, measured_height=50, # The MERRA dataset offers windspeeds at 50m target_height=TURBINE_HUB_HEIGHT, roughness=roughness, ) # 3 pressure_corrected_windspeeds = rk.wind.apply_air_density_adjustment( wind_speed=projected_windspeed, pressure=raw_pressure, temperature=raw_temperature, height=TURBINE_HUB_HEIGHT, ) convoluted_power_curve = power_curve.convolute_by_gaussian(scaling=0.06, base=0.1) # 4 power_curve = rk.wind.PowerCurve.from_capacity_and_rotor_diam( capacity=TURBINE_CAPACITY, rotor_diam=TURBINE_ROTOR_DIAMETER ) # 5 capacity_factors = convoluted_power_curve.simulate(wind_speed=pressure_corrected_windspeeds) # Print result capacity_factors

Out[9]:

	(6.25, 51.0)	(6.5, 51.0)	(6.25, 50.75)
2015-01-01 00:30:00+00:00	0.395655	0.265948	0.271238
2015-01-01 01:30:00+00:00	0.401460	0.266013	0.263638
2015-01-01 02:30:00+00:00	0.392434	0.262133	0.248822
2015-01-01 03:30:00+00:00	0.355571	0.242042	0.227111
2015-01-01 04:30:00+00:00	0.371400	0.258994	0.247654
...	...	...	...
2015-01-03 18:30:00+00:00	0.050925	0.006616	0.116428
2015-01-03 19:30:00+00:00	0.068637	0.054866	0.204342
2015-01-03 20:30:00+00:00	0.174980	0.163084	0.207436
2015-01-03 21:30:00+00:00	0.335079	0.337258	0.250530
2015-01-03 22:30:00+00:00	0.616502	0.553001	0.451090

71 rows × 3 columns

In [10]:

capacity_factors.plot()
plt.show()

capacity_factors.plot() plt.show()