Site specific turbine performance refers to the ability of a wind turbine to produce energy relative to a reference power curve given local wind conditions. A large fraction of the deviations from the reference power curve is due to the fact that the power curve is based on 10 minute average wind speed at hug-height (i.e. one location in space) with some reference values for wind shear and turbulence (typically a shear alfa on the order of 0.14 and a turbulence intensity of 10%). At different wind conditions the amount of energy over the swept rotor disc area is different for a given hub-height wind speed. The term “performance” is therefore somewhat misleading, as the majority of the “under-performance” of the wind turbine is simply due to the fact that there is less energy for the turbine to convert to electricity. There is also an effect that severe wind shear, wind veer and high turbulence cause difficult conditions which reduce the aerodynamic performance and thereby overall performance of the turbine, but this effect is secondary compared to the change in available energy in the air. The best news, however, is that both these effect can be simulated and accounted for.
Modern Energy has developed a method for calculating local turbine performance based on the transient wind regime at the turbine location. This is a major aspect of both layout design and yield analysis that is often overlooked by wind farm developers. The difference in local performance can vary by several percentage-points within a wind farm area, due to flow complexity caused by terrain or forest. Failing to considering local turbine performance in the layout process may therefore result in sub-optimal turbine layouts and over-predictions in annual yield estimates.
Our method is based on time series of measurement data from site. Turbine performance evolution needs to be based on transient calculations. The investigated effects are highly non-linear and intra-dependent, so both the distributions of the input parameters themselves needs to be considered, as does the covariance between the input parameters. Considering only time averaged properties always results in an over-prediction of turbine performance. To include the effect of local turbine performance in all subsequent analyses, we extract the local turbine performance to a scaled wind resource map which is later used for wind farm optimization and yield calculations.
Effects accounted for
Wind shear
Wind shear has two important effects which needs to be accounted for:
- The amount of energy available of the swept area varies with wind shear for a given hub-height wind speed. This may seem as an obvious statement, but as the expected production of a turbine is related to the hub-height wind speed and a reference wind shear via the power curve, this fact is often not properly accounted for in the wind industry.
- Wind shear also influence aerodynamic performance of the blades. This is due to periodic changes in inflow angle caused the change in oncoming wind speed as a function of height. This effect is small for moderate wind shears, but significant in large wind shear as it can cause periodic stall on the blades.
The largest error in turbine performance with respects to wind shear is typically experienced in areas when the wind shear is large at the lower half of the rotor, but small on the upper half of the rotor. In those situations both the energy available in the wind over both rotor halves are underestimated relative to reference power curve conditions.
Turbulence
The effect of turbulence also affect both the amount of energy available and turbine performance by the following processes:
- The time averaged amount of energy for a given mean wind speed increases with increasing turbulence intensity. This is due to the fact that the energy in the wind which passes a fixed location in space has a cubic relation to the instantaneous wind speed.
- Turbulence also affect blade aerodynamics, but this is typically a 2nd order effect compared to the energy aspect above.
Due to the shape of the power curve the turbine will produce more energy in low wind speeds, but less energy in wind speeds close to the rated power compared to a reference power curve when operating in high turbulence (and vice versa for low turbulence).
Dynamic yaw misalignment
Dynamic yaw misalignment is caused by fast changes in wind direction compared to the time required for a wind turbine controller to change direction. The time required before turbines yaw are turbine dependent, but is on the order of 60 seconds. The dynamic yaw misalignment reduces the energy produced by a turbine mainly due to a reduction in swept rotor area perpendicular to the wind direction.