Posters

Abstract

On the part of the wind industry, there is a great interest in increasing the time efficiency of wind farm planning. The central element here is the duration of a wind measurement campaign, which is usually one year. A reduction of the measurement duration has the potential to contribute decisively to an acceleration of yield estimation. In addition, a significantly higher cost efficiency can be achieved. Since only a smaller amount of data is available in such a case, specific challenges arise. Also, systematic seasonal biases often need to be addressed if the wind measurement does not cover all seasons. Up to now, no methods exist that take all this into account and guarantee a high accuracy even with reduced measurement time. This is where the STRAIGHT research project comes in. Procedures are being developed that make it possible to shorten the measurement period to, e.g., half a year. On the one hand, these procedures include ensuring a high-quality data basis. This primarily concerns the long-term data that are used for a long-term correction of the wind measurement (usually weather model or reanalysis data). For the long-term correction, new methods based on artificial intelligence are being developed that recognize and use seasonal patterns. The estimation of losses (often depending on the day/season) is also affected from a reduced measurement time and, therefore, is addressed. Further focal points of the project are the calculation of a site-specific power curve and the precise estimation of further design wind conditions (e.g., wind profile and turbulence). In detail, the following innovations and project goals are pursued: - Improvement of reanalysis data (based on the anemos wind atlas for Germany) - Development of a procedure to improve the measurement data basis (filling of data gaps) - Development of an AI-based procedure for the long-term correction of short-term wind measurements - Improvement of the loss calculation with regard to losses depending on time of day and season (e.g. animal protection, noise-reduced operation, shadow flicker, wakes) - Development of a site-specific correction of the power curve to account for effects such as turbulence or flow inclination on turbine yields - Improvement and use of geodata for the determination of local roughness and its change in time. In addition, methods for the objectified determination of all relevant uncertainties will be developed. In the end, the methods developed in the project will be combined in such a way that a highly automated estimation of the yields, including uncertainties, is achieved. The output is given in the form of reports. In addition, the methods are combined in such a way that yield estimation on a regional level is possible. This supports large-scale planning by both industry and politics. Members of the STRAIGHT project consortium are Fraunhofer IEE (project coordination), the University of Kassel, anemos GmbH and five other project partners from the wind industry.