Deriving multivariate probabilistic solar generation forecasts based on hourly imbalanced data

34 Pages Posted: 14 Jan 2025

See all articles by Yannik Pflugfelder

Yannik Pflugfelder

University of Duisburg-Essen - House of Energy Markets and Finance

Aiko Schinke-Nendza

University of Duisburg-Essen - House of Energy Markets and Finance

Jonathan Dumas

Réseau de Transport d'Electricité (RTE)

Christoph Weber

University of Duisburg-Essen

Date Written: November 29, 2024

Abstract

Accurate forecasting of solar PV generation is critical for integrating renewable energy into power systems. This paper presents a multivariate probabilistic forecasting model that addresses the challenges posed by imbalanced data resulting from day and night-time periods in solar photovoltaic (PV) generation. The proposed approach offers a robust and accurate method for predicting solar PV output by incorporating forecast updates and modeling the temporal interdependencies. The methodology is applied to a case study in France, demonstrating effectiveness across different spatial granularities and forecast horizons. The model uses advanced data handling methods combined with copula models, resulting in improved Energy Scores and Variogram-based Scores. These improvements underscore the importance of addressing imbalanced data and utilizing multivariate models with repeated updates to enhance solar forecasting accuracy. This work contributes to advancing forecasting techniques essential for integrating renewable energy into power grids, supporting the global transition to a sustainable energy future.

Keywords: Multivariate probabilistic forecasts, Forecast updates, Solar generation, Copula

Suggested Citation

Pflugfelder, Yannik and Schinke-Nendza, Aiko and Dumas, Jonathan and Weber, Christoph, Deriving multivariate probabilistic solar generation forecasts based on hourly imbalanced data (November 29, 2024). Available at SSRN: https://ssrn.com/abstract=5038630 or http://dx.doi.org/10.2139/ssrn.5038630

Yannik Pflugfelder (Contact Author)

University of Duisburg-Essen - House of Energy Markets and Finance ( email )

Lotharstrasse 1
Duisburg, 47048
Germany

Aiko Schinke-Nendza

University of Duisburg-Essen - House of Energy Markets and Finance ( email )

Lotharstrasse 1
Duisburg, 47048
Germany

Jonathan Dumas

Réseau de Transport d'Electricité (RTE) ( email )

Christoph Weber

University of Duisburg-Essen ( email )

Universitätsstraße 2
Essen, 45141
Germany

HOME PAGE: http://www.ewl.wiwi.uni-due.de

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
13
Abstract Views
84
PlumX Metrics