Modeling Wind Speed with a Long-Term Horizon and High-Time Interval with a Hybrid Fourier-Neural Network Model

The limited availability of local climatological stations and the limitations to predict wind speed (WS) accurately are significant barriers expansion energy (WE) projects worldwide. A methodology forecast WS at scale can be used overcome these barriers. This study proposes a with high-resolution long-term horizons, which combines Fourier model nonlinear autoregressive network (NAR). Given nonlinearities variations, NAR is based on variability identified analysis. modelled successfully 1.7 years wind-speed 3 hours time interval, what may considered longest forecasting horizon high resolution moment.