Solar irradiance prediction based on self-attention recursive model network

In recent years, with the continued development and popularity of sustainable energy sources increasing utilization solar energy, accurate radiation prediction has become important. this paper, we propose a new model based on deep learning, Feature-enhanced Gated Recurrent Unit, hereafter referred to as FEGRU, for prediction. This takes source data one-dimensional convolution self-attention feature attention processes features, then GRU performs extraction irradiance data. Finally, dimensionality is transformed by fully connected layer. The main advantage FEGRU that it does not require auxiliary data, but only time series can be used good Our experiments samples in Lyon, France, show our better results than baseline model.