Transformers Applied to Short-term Solar PV Power Output Forecasting

TL;DR

This paper explores the application of transformer architectures to short-term solar PV power output forecasting, comparing its performance with convolutional neural networks on image data over a year.

Contribution

It introduces a transformer-based model for solar PV output prediction and evaluates its effectiveness against existing CNN models using real-world data.

Findings

Transformer performs nearly as well as CNN in PV output prediction.

Transformer underperforms CNN on sunny days.

Model tuning with learning rate and batch size impacts results.

Abstract

Reliable forecasts of the power output from variable renewable energy generators like solar photovoltaic systems are important to balancing load on real-time electricity markets and ensuring electricity supply reliability. However, solar PV power output is highly uncertain, with significant variations occurring over both longer (daily or seasonally) and shorter (within minutes) timescales due to weather conditions, especially cloud cover. This paper builds on existing work that uses convolutional neural networks in the computer vision task of predicting (in a Nowcast model) and forecasting (in a Forecast model) solar PV power output (Stanford EAO SUNSET Model). A pure transformer architecture followed by a fully-connected layer is applied to one year of image data with experiments run on various combinations of learning rate and batch size. We find that the transformer architecture performs almost as well as the baseline model in the PV output prediction task. However, it performs worse on sunny days.