Christos Ioakimidis, Konstantinos Genikomsakis


The intermittent and unstable nature of wind raises significant challenges for the operation of wind power systems, either residential installations or utility-scale implementations, necessitating the development of reliable and accurate wind power forecasting techniques. Given that wind speed forecasting is typically considered the intermediate step for wind power forecasting, the present work proposes a novel short-term wind speed forecasting model based on an artificial neural network (ANN), with the key characteristic that statistical feature parameters of wind speed, wind direction and ambient temperature are employed in order to reduce the input vector and thus the complexity of the model. The results obtained indicate that the proposed model strikes a reasonable balance between accuracy and computational requirements for a forecasting time horizon of 24 hours, providing a light-weight solution that can be integrated as part of energy management systems for small scale applications.

Proceedings: Proceedings of the 41st Annual Conference of the IEEE Industrial Electronics Society
Presented at:

IECON (2015)





Christos Ioakimidis, Konstantinos Genikomsakis. (2015) "Short-term wind speed forecasting model based on ANN with statistical feature parameters" In Proceedings of the 41st Annual Conference of the IEEE Industrial Electronics Society. p. 971-976. DOI: 10.1109/IECON.2015.7392225.