24-h Ahead Wind Speed Prediction for the Optimum Operation of Hybrid Power Stations with the Use of Artificial Neural Networks
Remote areas are usually fed-in terms of electricity supply-from conventional generators that run on diesel. Recently, there is increasing interest on hybrid RES-based systems, including wind and solar power coupled with energy storage. To this end, optimum dispatching of such configurations is largely based on the capacity of prognostic tools employed in the respective energy management system. Acknowledging this, the aim of this work is the prediction of wind speed, 24-h ahead on an hourly basis, for the optimum operation of hybrid power stations (HPS) with the use of artificial neural networks (ANN). For this purpose, hourly data of wind speed have been used at a specific location (Tilos Island, Greece) where a HPS is going to be installed, including also a wind turbine of 800 kW. More specifically, an ANN which is fed with historical wind and air pressure data was developed in order to predict the wind speed at hub height on an hourly basis for the next 24 h. Results indicate that the proposed methodology gives an adequate forecast of wind speed in order to design an automated wind power information tool that could much facilitate the tasks of the energy management system.
KeywordsWind Speed Root Mean Square Error Artificial Neural Network Wind Turbine Artificial Neural Network Model
This work took place and was funded under the project TILOS (Horizon 2020 Low Carbon Energy Local/small-scale storage LCE-08-2014). This project has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement No 646529. Open image in new window
- Nastos P, Moustris K, Larissi I, Paliatsos A (2011) Air quality and bioclimatic conditions within the greater athens area, Greece-development and applications of artificial neural networks. Adv Air Pollut, 557–584. InTech-Open Access (ISBN: 978-953-307-511-2)Google Scholar
- Nielsen TS, Madsen H, Nielsen HAa, Pinson P, Kariniotakis G, Siebert N, Marti I, Lange M, Focken U, Lueder V, Bremen LV, Louka P, Kallos G, Galanis G (2006) Short-term wind power forecasting using advanced statistical methods. In: The European wind energy conference, EWEC 2006, Athens, Greece, 9–17 Feb 2006Google Scholar