Long-term Solar Radiation Forecasting using a Deep Learning Approach-GRUs

A long-term solar generation forecasting is an important issue in a microgrid design. Solar generation forecasting mainly depends on solar radiation forecasting. In this paper, A Deep Learning approach using Gated Recurrent Units (GRUs) is proposed for forecasting of a year-ahead hourly and daily solar radiation. The proposed GRU model is compared with the state of the art methods like Long Short Term Memory (LSTM), Recurrent Neural Network (RNN), Support Vector Regression (SVR), Feed Forward Neural Network (FFNN) models and numerical model. Its effectiveness for long-term solar radiation forecasting over other methods is verified.