To main content

Deep Learning-Based Solar Irradiance Decomposition Models for Nordic Regions

Abstract

This work presents a comparative evaluation of machine learning (ML) and deep learning (DL) models for solar irradiance decomposition in Nordic regions, where traditional empirical models often struggle. Using data from the Alpha Centauri outdoor test facility in Trondheim, Norway, the present work benchmarks the performance of Histogram-based Gradient Boosting (HGB), Artificial Neural Networks (ANN), and Long Short-Term Memory (LSTM) networks against the Erbs model. Results indicate that HGB performs best on the initial evaluation set, achieving strong R² scores for both DNI and DHI, while requiring minimal computational resources. However, in a zero-shot prediction scenario using independent data, HGB's performance drops significantly, suggesting overfitting to seasonal patterns. ANN maintains the highest accuracy for DNI decomposition, capturing nonlinear dependencies more effectively, whereas LSTM shows mixed results, particularly underperforming in DNI estimation.

Category

Conference poster

Language

English

Affiliation

  • SINTEF Industry / Sustainable Energy Technology

Presented at

42nd European Photovoltaic Solar Energy Conference and Exhibition

Place

Bilbao

Date

22.09.2025 - 26.12.2025

Organizer

EUPVSEC

Year

2025

View this publication at Norwegian Research Information Repository