Using sky-classification to improve the short-term prediction of irradiance with sky images and convolutional neural networks

Using sky-classification to improve the short-term prediction of irradiance with sky images and convolutional neural networks

Martinez Lopez, V.A. (TU Delft Photovoltaic Materials and Devices; TU Delft Electrical Engineering, Mathematics and Computer Science)
van Urk, G.A. (TU Delft Photovoltaic Materials and Devices; TU Delft Electrical Engineering, Mathematics and Computer Science)
Doodkorte, P.J.F. (TU Delft Photovoltaic Materials and Devices; TU Delft Electrical Engineering, Mathematics and Computer Science)
Zeman, M. (TU Delft Photovoltaic Materials and Devices)
Isabella, O. (TU Delft Photovoltaic Materials and Devices)
Ziar, H. (TU Delft Photovoltaic Materials and Devices)

Electrical Engineering, Mathematics and Computer Science

2024

Clouds moving in front or away from the sun are the leading cause of irradiance variability. These variations have a repercussion on the electricity production of photovoltaic systems. Predicting such changes is essential for proper control of these systems and for maintaining grid stability. Images from the sky have proven to help with short-term solar irradiance forecasting, especially when combined with artificial intelligence. Nevertheless, these models tend to smooth the irradiance fluctuations. We propose a forecasting model to predict the clear-sky index in a forecast horizon of 20 min with a 1-minute resolution. Our model, based on a classifier to determine the sky conditions and, on an optical flow, applies an artificial intelligence model explicitly trained on each class of sky conditions. This strategy has an equivalent performance to an unclassified model and a forecast skill between 5 and 20% with respect to the smart persistence model for most classes of sky conditions while requiring considerably less training data. Although our model reduces the overall predicting error, it still has difficulties predicting irradiance changes and mainly overcast days. Our classifying strategy can be applied to other models targeting different objectives to predict sudden changes in either irradiance or power related to photovoltaic systems.

All-sky images
Deep learning
Irradiance nowcasting
Sky-image processing

http://resolver.tudelft.nl/uuid:d92ada99-8261-46c1-ac6f-de2e3feb3520

https://doi.org/10.1016/j.solener.2024.112320

0038-092X

Solar Energy, 269

Institutional Repository

journal article

© 2024 V.A. Martinez Lopez, G.A. van Urk, P.J.F. Doodkorte, M. Zeman, O. Isabella, H. Ziar