Authors:
Jane Oktavia Kamadinata
1
;
Tan Lit Ken
1
and
Tohru Suwa
2
Affiliations:
1
Universiti Teknologi Malaysia, Malaysia
;
2
President University, Indonesia
Keyword(s):
Artificial Neural Network, Global Solar Radiation Prediction, Sky Image, Photovoltaic Power Generation.
Related
Ontology
Subjects/Areas/Topics:
Energy and Economy
;
Energy-Aware Systems and Technologies
;
Renewable Energy Resources
Abstract:
Solar radiation is an essential source of energy that has yet to be fully utilized. This energy can be converted into another form of more usable energy, electricity, by using photovoltaic power generation systems in order to fight against global warming. When the photovoltaic power generation systems are connected to an electrical grid, predicting near-future global solar radiation is important to stabilize the entire network. Two different simple methodologies utilizing artificial neural networks (ANNs) to predict the global solar radiation in 1 to 5 minutes in advance from sky images are developed and compared. In the first methodology, two ANNs are combined. The first ANN predicts cloud movement direction, while the second ANN predicts global solar radiation using the first ANN’s prediction results. On the other hand, a single ANN directly predicts global solar radiation in the second methodology. Both of the proposed methodologies are able to capture the trends of the global sol
ar radiation well. Because the proposed methodologies only use limited number of sampling points, the computational effort is significantly reduced compared to the existing methodologies where the whole images need processing.
(More)