Short-range wind speed predictions for complex terrain using an interval-artificial neural network

Abstract

Renewable energy such as wind and solar energy rely on favorable weather conditions. As load balancing in the energy system is crucial accurate and tailored forecasts of the expected wind speed and power production are needed. A neural network (NN) based approach for short-range wind speed forecasts was developed. Two different NN models were developed using observations and NWP data as input. The interval-based NN (iANN) approach outperformed the NWP models and a MOS-based forecast and was able to reproduce the observations at 25 representative Austrian observation sites.