Skip to main content

Solar Irradiance Anticipative Transformer.

Mercier, T. M., Rahman, T. and Sabet, A., 2023. Solar Irradiance Anticipative Transformer. In: O’Conner, L., ed. 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). New York, NY: IEEE, 2065-2074.

Full text available as:

Mercier_Solar_Irradiance_Anticipative_Transformer_CVPRW_2023_paper.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial.


DOI: 10.1109/CVPRW59228.2023.00200


This paper proposes an anticipative transformer-based model for short-term solar irradiance forecasting. Given a sequence of sky images, our proposed vision transformer encodes features of consecutive images, feeding into a transformer decoder to predict irradiance values associated with future unseen sky images. We show that our model effectively learns to attend only to relevant features in images in order to forecast irradiance. Moreover, the proposed anticipative transformer captures long-range dependencies between sky images to achieve a forecasting skill of 21.45 % on a 15 minute ahead prediction for a newly introduced dataset of all-sky images when compared to a smart persistence model.

Item Type:Book Section
Additional Information:Conference Location: Vancouver, BC, Canada - Date of Conference: 17-24 June 2023
Uncontrolled Keywords:Training; Computer vision; Conferences; Predictive models; Feature extraction; Transformers; Decoding; feature extraction; learning (artificial intelligence); load forecasting; neural nets; power engineering computing; solar power; all-sky images; anticipative transformer-based model; anticipative transformers; consecutive images; irradiance values; short-term solar irradiance forecasting; sky images; smart persistence model; solar irradiance anticipative transformer; time 15.0 min; transformer decoder; vision transformer
Group:Faculty of Science & Technology
ID Code:39066
Deposited By: Symplectic RT2
Deposited On:16 Oct 2023 12:15
Last Modified:16 Oct 2023 12:17


Downloads per month over past year

More statistics for this item...
Repository Staff Only -