Definitions of key terms used in wind resource assessment, turbine technology, and wind farm development.
A
AEP (Annual Energy Production)
The total amount of electrical energy a wind farm is expected to generate over a one-year period, measured in megawatt-hours (MWh) or gigawatt-hours (GWh). AEP is calculated by multiplying the installed capacity by the capacity factor and by 8,760 hours per year. WindAI returns AEP estimates as part of every assessment, derived from predicted hourly capacity factors across all 8,760 hours of the year.
Availability
The percentage of time a wind turbine or wind farm is operational and able to generate electricity when wind conditions are suitable. Modern wind turbines typically achieve availability rates of 95-98%. Availability losses include scheduled maintenance, unplanned outages, and grid curtailment. Net capacity factor calculations account for availability reductions.
B
Boundary Layer Height
The depth of the atmospheric boundary layer, which is the lowest part of the atmosphere directly influenced by the Earth's surface. Boundary layer height varies from a few hundred meters at night to 1-3 km during the day over land, and affects vertical wind profiles and turbulence characteristics. WindAI uses ERA5 boundary layer height data as one of its 391 input features to improve capacity factor predictions.
C
Capacity Factor
The ratio of a wind farm's actual energy output over a period to the maximum possible output if it operated at full rated power continuously. Expressed as a percentage, typical onshore capacity factors range from 25-45%, while offshore farms can reach 40-55%. WindAI predicts capacity factors using a deep neural network trained on 10M+ hourly observations from 289 wind farms, achieving an RMSE of 0.1477 and R-squared of 0.777.
Cut-in Speed
The minimum wind speed at which a wind turbine begins generating electricity, typically between 2.5 and 4 m/s depending on the turbine model. Below the cut-in speed, the energy in the wind is insufficient to overcome the turbine's mechanical and electrical losses. Cut-in speed is a key parameter on the turbine power curve.
Cut-out Speed
The wind speed at which a wind turbine shuts down to protect itself from structural damage, typically between 25 and 34 m/s. When wind speeds exceed the cut-out threshold, the turbine's braking system engages and the blades are feathered to minimize aerodynamic loads. Some modern turbines use storm ride-through or soft cut-out strategies to reduce production losses during high-wind events.
D
Diurnal Pattern
The daily cycle of wind speed variation caused by solar heating and cooling of the Earth's surface. In many locations, winds are stronger during the afternoon when thermal convection is greatest and weaker at night when the atmosphere stabilizes. WindAI captures diurnal patterns through temporal encoding features and returns hourly predictions that reflect these daily cycles.
Direct Drive
A wind turbine drivetrain configuration that connects the rotor directly to the generator without a gearbox. Direct drive turbines use low-speed, high-torque generators (typically permanent magnet synchronous generators) and eliminate gearbox-related maintenance and failures. Major direct drive platforms include the Siemens Gamesa SG 14-236 DD and GE Haliade-X.
E
ERA5
The fifth-generation atmospheric reanalysis dataset produced by the European Centre for Medium-Range Weather Forecasts (ECMWF), providing hourly estimates of atmospheric variables on a 31 km grid from 1940 to near-present. ERA5 includes wind speed, wind direction, temperature, pressure, and boundary layer parameters at multiple pressure levels. WindAI uses ERA5 data as its primary atmospheric input, extracting 6 variables across 16 grid points surrounding each target location.
Exceedance Probability (P50, P75, P90, P99)
A statistical measure indicating the probability that annual energy production will exceed a given value. P50 means there is a 50% chance production will exceed the stated value (the median estimate). P90 means there is a 90% probability of exceeding the value, representing a conservative estimate that lenders typically require for debt sizing. WindAI reports both P50 and P90 values for every assessment.
G
GELU (Gaussian Error Linear Unit)
An activation function used in neural networks that smoothly approximates ReLU by weighting inputs by their percentile under a Gaussian distribution. GELU is used in WindAI's 6-layer neural network architecture, providing smoother gradients during training compared to ReLU. It has become the preferred activation function in modern deep learning architectures including transformers.
Grid Connection
The electrical infrastructure required to connect a wind farm to the power transmission or distribution network. Grid connection costs can represent 10-20% of a wind farm's total capital expenditure and include substations, transformers, and transmission lines. The availability and capacity of grid connection points significantly influences wind farm site selection.
H
Hub Height
The height above ground level of the center of a wind turbine's rotor, where the nacelle and generator are mounted. Modern onshore turbines typically have hub heights of 80-170 meters, with taller towers accessing stronger and more consistent winds. Hub height is a critical input for energy production estimates because wind speed generally increases with height according to the wind shear profile.
I
IEC Wind Class (I, II, III, IV)
A classification system defined by the International Electrotechnical Commission (IEC 61400-1) that categorizes wind turbines based on the wind conditions they are designed to withstand. Class I turbines are designed for high-wind sites (annual average >8.5 m/s at hub height), Class II for medium-wind (7.5-8.5 m/s), and Class III for low-wind (<7.5 m/s). Selecting the correct IEC class for a site is essential for structural safety and optimal energy capture.
Installed Capacity
The total maximum power output of a wind farm under ideal conditions, calculated as the number of turbines multiplied by the rated power of each turbine, measured in megawatts (MW). A wind farm with 50 turbines of 4 MW each has an installed capacity of 200 MW. Installed capacity represents the theoretical maximum, while actual production depends on the capacity factor.
L
LCOE (Levelized Cost of Energy)
The total lifetime cost of building and operating a power plant divided by its total energy output, expressed in dollars per megawatt-hour ($/MWh). LCOE enables comparison across different generation technologies on a common basis. Global average LCOE for onshore wind was approximately $30-50/MWh in 2025, making it one of the cheapest sources of new electricity generation in most markets.
M
Mean Wind Speed
The average wind speed at a given location and height over a specified time period, typically expressed in meters per second (m/s). Long-term mean wind speed is one of the most fundamental parameters in wind resource assessment, though it alone is insufficient for accurate energy predictions because the shape of the wind speed distribution also matters significantly. Sites with the same mean wind speed can have different capacity factors depending on wind speed variability.
MERRA-2
The Modern-Era Retrospective Analysis for Research and Applications, Version 2 — a global atmospheric reanalysis dataset produced by NASA's Global Modeling and Assimilation Office. MERRA-2 provides hourly data at approximately 50 km resolution from 1980 to present. WindAI incorporates MERRA-2 data alongside ERA5 to improve prediction robustness, particularly for boundary layer height estimates.
N
Nacelle
The housing at the top of a wind turbine tower that contains the key generating components: the gearbox (if present), generator, yaw system, and control electronics. Modern utility-scale nacelles weigh 50-400 tonnes depending on the turbine platform. Nacelle-mounted anemometers are commonly used for operational wind speed measurement, though they are affected by rotor wake.
Net Capacity Factor
The capacity factor after accounting for all real-world losses including wake effects, electrical losses, turbine availability, environmental curtailment, and grid curtailment. Net capacity factor is always lower than gross capacity factor and is the metric used for financial modeling and PPA negotiations. A typical onshore wind farm might have a gross capacity factor of 35% and a net capacity factor of 30% after all losses.
O
Offshore Wind
Wind energy generated by turbines installed in bodies of water, typically on the continental shelf. Offshore wind farms benefit from stronger, more consistent winds and fewer terrain-induced turbulence effects, achieving capacity factors of 40-55% compared to 25-45% for onshore. Offshore installation costs are 2-3 times higher than onshore, but declining rapidly with fixed-bottom and floating foundation technologies.
Onshore Wind
Wind energy generated by turbines installed on land. Onshore wind is the most mature and cost-effective form of wind energy, with global installed capacity exceeding 900 GW. Typical onshore capacity factors range from 25-45% depending on site quality, turbine technology, and wind resource. WindAI's training dataset includes onshore wind farms across 8 countries representing diverse terrain and climate conditions.
P
Power Curve
A graph or table showing the relationship between wind speed and the electrical power output of a wind turbine. The power curve defines four key regions: zero output below cut-in speed, a cubic power increase in the partial load region, rated power output between rated speed and cut-out speed, and zero output above cut-out speed. Power curve data from manufacturers is essential for converting wind speed distributions into energy production estimates.
Power Density
The amount of wind energy available per unit area of rotor swept area, measured in watts per square meter (W/m2). Power density depends on wind speed cubed and air density, making it a more informative metric than mean wind speed alone. A site with 400-500 W/m2 at hub height is generally considered to have an excellent wind resource for commercial development.
R
Rated Power
The maximum electrical power output a wind turbine is designed to produce, typically achieved at the rated wind speed (usually 11-15 m/s). Rated power is specified in megawatts (MW) and defines the nameplate capacity of the turbine. Modern onshore turbines range from 3-7 MW rated power, while offshore turbines reach 12-18 MW.
Rated Wind Speed
The wind speed at which a wind turbine first reaches its maximum (rated) power output, typically between 11 and 15 m/s. Above rated speed, the turbine's pitch control system feathers the blades to maintain constant power output and prevent overloading. Lower rated speeds generally indicate turbines optimized for lower-wind sites.
Rotor Diameter
The diameter of the circle swept by a wind turbine's blades, measured in meters. Rotor diameter is the primary driver of energy capture because swept area increases with the square of the diameter. Modern onshore turbine rotors range from 120-175 meters, while the largest offshore rotors exceed 230 meters. Larger rotors capture more energy but increase structural loads and transportation logistics.
Roughness Length
A parameter that characterizes the aerodynamic roughness of the terrain surface and its effect on wind speed profiles, measured in meters. Smooth surfaces like open water have roughness lengths of 0.0001-0.001 m, grassland around 0.01-0.05 m, and forests or urban areas 0.5-2.0 m. Higher roughness reduces wind speed near the surface and increases wind shear, making roughness length a critical input for vertical wind speed extrapolation.
S
Shear Exponent
A dimensionless parameter in the power law wind profile equation that describes how wind speed changes with height above ground. The power law states that V2/V1 = (H2/H1)^alpha, where alpha is the shear exponent. Typical values range from 0.10 over open water to 0.25 over forested terrain. Accurate shear estimation is important because small changes in the exponent significantly affect energy production estimates at hub height.
Specific Power
The ratio of a wind turbine's rated power to its rotor swept area, measured in watts per square meter (W/m2). Lower specific power (200-300 W/m2) indicates a larger rotor relative to generator size, producing higher capacity factors at lower wind speeds. Higher specific power (350-450 W/m2) is suited to high-wind sites. Specific power is one of the most important turbine selection parameters for site optimization.
Swept Area
The circular area swept by a wind turbine's rotating blades, calculated as pi times the rotor radius squared. Swept area directly determines how much kinetic energy the turbine can extract from the wind. A turbine with a 150-meter rotor diameter has a swept area of approximately 17,671 square meters. Doubling the rotor diameter quadruples the swept area and approximately quadruples energy capture.
T
Turbulence Intensity
The ratio of the standard deviation of wind speed fluctuations to the mean wind speed over a defined period (typically 10 minutes), expressed as a percentage. Higher turbulence increases structural fatigue loads on turbine components and can reduce energy capture. IEC standards define turbulence categories (A, B, C) that determine turbine design requirements. Complex terrain and wake effects both increase turbulence intensity.
Turbine Availability
The percentage of time a wind turbine is technically capable of generating electricity, excluding periods of maintenance, repair, and unplanned outages. Modern turbines achieve technical availability of 95-98%. Contractual availability guarantees from manufacturers typically specify 95-97% as the minimum threshold, with penalties for underperformance.
W
Wake Effect
The reduction in wind speed and increase in turbulence downstream of a wind turbine, caused by the extraction of kinetic energy from the airflow. Wake effects can reduce energy production of downstream turbines by 5-20% depending on spacing, layout, and atmospheric conditions. Wind farm layout optimization aims to minimize wake losses while maximizing the number of turbines within the available land area.
Weibull Distribution
A two-parameter probability distribution commonly used to describe the frequency distribution of wind speeds at a given location. The shape parameter (k) describes the variability of wind speeds, while the scale parameter (A) is related to the mean wind speed. Typical shape parameters for wind sites range from 1.5 to 3.0, with higher values indicating more consistent winds. The Weibull distribution is fundamental to wind resource assessment and energy yield calculations.
Wind Rose
A circular diagram showing the frequency and intensity of wind from different directions at a specific location. Wind roses typically display 12 or 16 directional sectors and use color coding or concentric rings to indicate wind speed ranges. Understanding the prevailing wind direction is essential for wind farm layout optimization, as turbines must be spaced to minimize wake effects along the dominant wind axis.
Wind Shear
The change in wind speed or direction with height above the ground surface. Vertical wind shear causes the wind speed at the top of the rotor to differ from the speed at the bottom, creating uneven loading across the swept area. Wind shear is influenced by surface roughness, atmospheric stability, and time of day. Accurate characterization of wind shear is critical for extrapolating measured or modeled wind speeds to the turbine hub height.
Y
Yaw System
The motorized mechanism that rotates the nacelle and rotor of a wind turbine to face into the wind. Modern yaw systems use electric motors and a large slewing bearing to continuously adjust the turbine's orientation based on wind direction measurements from nacelle-mounted sensors. Yaw misalignment of even a few degrees can reduce energy capture by 1-3%, making accurate yaw control important for maximizing production.
Put these concepts into practice
Get AI-powered wind resource assessments that calculate capacity factor, AEP, P50/P90, and more for any location on Earth.
