The wind power conversion model estimates wind power generation potential globally, using gridded meteorological inputs and detailed turbine specifications. Designed for flexibility, the model can be applied to various data streams—including historical reanalysis (e.g. ERA5), seasonal forecasts, and climate projections—and supports both onshore and offshore wind domains. Output includes gridded and aggregated capacity factor (CF) datasets, enabling downstream energy modelling and climate-energy integration.

Input Data

The model requires as input:

• Gridded bias-adjusted wind speed data (NetCDF format), typically at hub heights (e.g., 100 m) or at different reference heights, depending on the turbine configuration.

• Turbine characteristics, including power curves, hub height, rotor diameter, and design class (onshore/offshore), from TheWindPower.net, International Energy Agency (IEA) and National Renewable Energy Laboratory (NREL).

• Turbine placement and capacity (optional for regional aggregation), to convert capacity factors into generation estimates.

Model Description

The WP model operates by applying wind turbine power curves to wind speed data, extrapolated to the turbine hub height if necessary. The procedure includes:

1. Wind Speed Extrapolation: If wind speed is provided at a reference height different from the turbine hub height, it is extrapolated using the Power Law for Wind Profile Scaling.
   As a rule:

   • If hub height < 100 m: use 10 m as reference

   • If hub height > 100 m: use 100 m as reference

   • If hub height = reference: no extrapolation is applied

   Figure 2.1 shows the difference in wind speeds between 150 m and 100 m as extrapolated using the power law, highlighting the relevance of height correction for tall turbines.

2. Power Curve Application: For each turbine and time step, the extrapolated wind speed is mapped onto a power output using the turbine's power curve (Figure 2.2).

3. Capacity Factor Calculation: The power output is converted to capacity factor (CF) by normalising the generated power

   P_{gen}

   ​ by the rated (installed) capacity

   P_{rated}

   ​ of the turbine:

CF = \frac{P_{gen}}{P_{rated}}

This yields time series of CF for each grid cell and turbine type.

Figure 2.1: Global difference in wind speeds at 150 m and 100 m hub heights using the power law.

Figure 2.2: Power curves for selected turbines: current standards (V164/8000, 2.5-103 AND G132/3300) and future scenarios (15MW_240_RWT and spoke_6_MW_170).

Turbine Specifications

The wind power model uses a set of representative onshore and offshore turbines. For turbines with hub heights ≥ 100 m, two capacity-factor versions are provided in the historical dataset because two different 100 m wind speed products are available. Table 2.1 below lists the selected turbines and specifies which wind-speed input is used for each version.

Table 2.1: Selected turbine types used in the Wind Power model, showing their main technical parameters and the corresponding input wind-speed sources. The corresponding power curves are plotted in Figure 2.2. For turbines with hub heights ≥ 100 m, two capacity-factor versions are provided, based on the two WS100 products available in the historical dataset. 

*Note: in the historical stream and for turbines with hub heights ≥ 100 m (turbines # 0, 1, 3, 4), two versions of the wind power capacity factor are provided. 
These correspond to the two available WS100 products:

1. WS100 estimated from bias-adjusted WS10 using the power-law/Alpha formulation;

2. WS100 computed from ERA5 and bias-adjusted with GWA2.

Output Data

• Hourly gridded (NetCDF) wind power capacity factors per turbine type
• For the onshore wind power indicator (WON), after applying the Spatial Aggregation Procedure and the Temporal Aggregation Procedure, aggregated datasets are computed at the ADM0 and ADM1 levels for hourly, daily, monthly, seasonal, and annual temporal resolutions.

Related articles

Related articles appear here based on the labels you select. Click to edit the macro and add or change labels.