Estimation of Global Solar Radiation from Air Temperature Using Artificial Neural Networks Based on Reservoir Computing
Abstract
Solar radiation data are crucial for the design and evaluation of solar energy systems. For locations where measurements are not available, models to estimate solar radiation from more readily available data are required. In this paper, we introduce the use of the reservoir computing (RC) technique to model daily global solar radiation (GSR) as a function of air temperature. RC is a type of artificial neural network (ANN) with closed loops (recurrent neural network, RNN) particularly suited to process time-dependent information. The novelty of this work is the use of a recurrent network for modeling solar radiation, which allows including temporal correlations between the input and output variables. The proposed approach is used to estimate GSR using weather data from a location in Almeria, Spain. The results show higher accuracy than those obtained with conventional regression models.
Keywords
Global solar radiation, Modeling, Artificial neural networks, Reservoir computing