Optimisation of design and operational management for hybrid power plants and energy storage technologies by means of wind and PV power nowcasting
- Nowcasting of wind power feed-in of individual wind turbines (WT) and of wind farms by combining high temporal resolution long-range lidar measurements at a wind turbine with further meteorological data. The combination of the data and derivation of predictions will be carried out with a state-of-the-art machine learning method. Machine learning methods will also be developed to improve persistence forecasting in order to extend the time domain of the lidar measurements as much as possible.
- Nowcasting of PV power with high spatial and temporal resolution for regions. This involves on-site measurements with a cloud camera with high temporal resolution (<=1 minute) combined with satellite data (temporal resolution of 15 minutes). Predictions are generated with the help of machine learning methods.
- Extension of the P2IONEER model to enable the use of nowcasting. The model is extended to allow the use of nowcasting and is thus developed into a universal tool for the design, optimization and operation of combined renewable energy power plants. The operators and technical interfaces will be adapted in order to automate use of the model as far as possible. The coupling of power and gas grids and the resulting reduced grid loads will be closely investigated.
VORKAST is lead by the Centre for Solar Energy and Hydrogen Research Baden-Württemberg (ZSW).
- VORKAST (11/13/2017)
- I. Würth, M. Wigger, P.W. Cheng (2016). “Nowcasting the Power Output of a Wind Turbine using a Long Range Lidar”. Presented at ISARS 2016, Varna, Bulgaria. Download