Solar Irradiance Anticipative Transformer

TL;DR

This paper introduces an anticipative transformer model that leverages sky image sequences to accurately forecast short-term solar irradiance, demonstrating significant improvements over baseline models.

Contribution

The paper presents a novel vision transformer architecture specifically designed for anticipative solar irradiance forecasting using sky images.

Findings

Achieved 21.45% forecasting skill on 15-minute ahead predictions.

Effectively captures long-range dependencies in sky image sequences.

Outperforms a smart persistence baseline model.

Abstract

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.