An Evaluation of Predictor Variables for Photovoltaic Power Forecasting

Lennard Visser*, Tarel AlSkaif, Wilfried van Sark

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference paperAcademicpeer-review

1 Citation (Scopus)

Abstract

Accurate forecasts of the electric power generation by solar Photovoltaic (PV) systems are essential to support their vast increasing integration. This study evaluates the interdependence of 14 predictor variables and their importance to machine learning models that forecast the day-ahead PV power production. To this purpose, we use two feature selection models to rank the predictor variables and accordingly, examine the performance change of two forecast models when a growing number of variables is considered. The study is performed using 3 years of data for Utrecht, the Netherlands. The results show the most important variables for PV power forecasting and identifies how many top variables should be considered to achieve an optimal forecast performance accuracy. Additionally, the best forecast model performance is found when only a few predictor variables are considered, including a created variable that estimates the PV power output based on technical system characteristics and physical relations.

Original languageEnglish
Title of host publicationIntelligent Technologies and Applications - 4th International Conference, INTAP 2021, Revised Selected Papers
EditorsFilippo Sanfilippo, Ole-Christoffer Granmo, Sule Yildirim Yayilgan, Imran Sarwar Bajwa
Place of PublicationCham
PublisherSpringer
Pages303-310
Number of pages8
ISBN (Print)9783031105241
DOIs
Publication statusPublished - 2022
Event4th International Conference on Intelligent Technologies and Applications, INTAP 2021 - Grimstad, Norway
Duration: 11 Oct 202113 Oct 2021

Publication series

NameCommunications in Computer and Information Science
Volume1616 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference/symposium

Conference/symposium4th International Conference on Intelligent Technologies and Applications, INTAP 2021
Country/TerritoryNorway
CityGrimstad
Period11/10/2113/10/21

Keywords

  • Machine learning
  • Photovoltaics
  • Predictor variables
  • Solar power forecasting
  • Weather forecasts

Fingerprint

Dive into the research topics of 'An Evaluation of Predictor Variables for Photovoltaic Power Forecasting'. Together they form a unique fingerprint.

Cite this