A very simple gust estimation algorithm for wind energy applications

Rapid changes of wind speed, wind gusts, can harm the structural safety of wind turbines, reduce their energy output, lead to a shorter rotor blade life cycle due to fatigue, and reduce the lifetime of other wind turbine components. Besides damages to infrastructure, the oscillations and / or ramping events can result in a fast fluctuation of the grid voltage and add additional burden to the power grid.

Thus, accurate forecasting of wind gusts and gustiness for power grid balancing, trading, can become a crucial factor. However, often no wind gust measurements are available from the wind farm operators (SCADA, mast if available) as only 10-minute wind speed averages can be used. Often, the environmental conditions surrounding a wind farm, e.g. roughness length, are unknown, too. Thus, a very simple gust estimation algorithm based on the 10-minute SCADA wind speed averages is needed.

In the past, several attempts have been made to estimate the wind gusts based on very simple assumptions (Wieringa, 1973, Cvitan, 2004, Harper et al., 2010). However, those algorithms all have their flaws.

To account for gustiness a very simple gust estimation algorithm was developed and tested for different heights above ground and locations (flat, mountainous), and observation qualities such as standard synop-sites, mast measurements, and SCADA data. The algorithm was developed in a way that it can be used in a deterministic and probabilistic setup. A verification of the algorithm against measurements showed that it is able to outperform other, similarly simple algorithms and that it is able to provide a good estimate of gustiness for those locations.