Using weather data in energy time series forecasting: the benefit of input data transformations

Abstract Renewable energy systems depend on the weather, and weather information, thus, plays a crucial role in forecasting time series within such renewable energy systems. However, while weather data are commonly used to improve forecast accuracy, it still has to be determined in which input shape...

Full description

Bibliographic Details
Main Authors: Oliver Neumann, Marian Turowski, Ralf Mikut, Veit Hagenmeyer, Nicole Ludwig
Format: Article
Language:English
Published: SpringerOpen 2023-11-01
Series:Energy Informatics
Subjects:
Online Access:https://doi.org/10.1186/s42162-023-00299-8
_version_ 1797636624294608896
author Oliver Neumann
Marian Turowski
Ralf Mikut
Veit Hagenmeyer
Nicole Ludwig
author_facet Oliver Neumann
Marian Turowski
Ralf Mikut
Veit Hagenmeyer
Nicole Ludwig
author_sort Oliver Neumann
collection DOAJ
description Abstract Renewable energy systems depend on the weather, and weather information, thus, plays a crucial role in forecasting time series within such renewable energy systems. However, while weather data are commonly used to improve forecast accuracy, it still has to be determined in which input shape this weather data benefits the forecasting models the most. In the present paper, we investigate how transformations for weather data inputs, i. e., station-based and grid-based weather data, influence the accuracy of energy time series forecasts. The selected weather data transformations are based on statistical features, dimensionality reduction, clustering, autoencoders, and interpolation. We evaluate the performance of these weather data transformations when forecasting three energy time series: electrical demand, solar power, and wind power. Additionally, we compare the best-performing weather data transformations for station-based and grid-based weather data. We show that transforming station-based or grid-based weather data improves the forecast accuracy compared to using the raw weather data between 3.7 and 5.2%, depending on the target energy time series, where statistical and dimensionality reduction data transformations are among the best.
first_indexed 2024-03-11T12:37:44Z
format Article
id doaj.art-7fe36b276171407f92a995d9ef748f9e
institution Directory Open Access Journal
issn 2520-8942
language English
last_indexed 2024-03-11T12:37:44Z
publishDate 2023-11-01
publisher SpringerOpen
record_format Article
series Energy Informatics
spelling doaj.art-7fe36b276171407f92a995d9ef748f9e2023-11-05T12:30:53ZengSpringerOpenEnergy Informatics2520-89422023-11-016112310.1186/s42162-023-00299-8Using weather data in energy time series forecasting: the benefit of input data transformationsOliver Neumann0Marian Turowski1Ralf Mikut2Veit Hagenmeyer3Nicole Ludwig4Institute for Automation and Applied Informatics, Karlsruhe Institute of TechnologyInstitute for Automation and Applied Informatics, Karlsruhe Institute of TechnologyInstitute for Automation and Applied Informatics, Karlsruhe Institute of TechnologyInstitute for Automation and Applied Informatics, Karlsruhe Institute of TechnologyCluster of Excellence Machine Learning, University of TübingenAbstract Renewable energy systems depend on the weather, and weather information, thus, plays a crucial role in forecasting time series within such renewable energy systems. However, while weather data are commonly used to improve forecast accuracy, it still has to be determined in which input shape this weather data benefits the forecasting models the most. In the present paper, we investigate how transformations for weather data inputs, i. e., station-based and grid-based weather data, influence the accuracy of energy time series forecasts. The selected weather data transformations are based on statistical features, dimensionality reduction, clustering, autoencoders, and interpolation. We evaluate the performance of these weather data transformations when forecasting three energy time series: electrical demand, solar power, and wind power. Additionally, we compare the best-performing weather data transformations for station-based and grid-based weather data. We show that transforming station-based or grid-based weather data improves the forecast accuracy compared to using the raw weather data between 3.7 and 5.2%, depending on the target energy time series, where statistical and dimensionality reduction data transformations are among the best.https://doi.org/10.1186/s42162-023-00299-8Energy time seriesForecastingWeather data
spellingShingle Oliver Neumann
Marian Turowski
Ralf Mikut
Veit Hagenmeyer
Nicole Ludwig
Using weather data in energy time series forecasting: the benefit of input data transformations
Energy Informatics
Energy time series
Forecasting
Weather data
title Using weather data in energy time series forecasting: the benefit of input data transformations
title_full Using weather data in energy time series forecasting: the benefit of input data transformations
title_fullStr Using weather data in energy time series forecasting: the benefit of input data transformations
title_full_unstemmed Using weather data in energy time series forecasting: the benefit of input data transformations
title_short Using weather data in energy time series forecasting: the benefit of input data transformations
title_sort using weather data in energy time series forecasting the benefit of input data transformations
topic Energy time series
Forecasting
Weather data
url https://doi.org/10.1186/s42162-023-00299-8
work_keys_str_mv AT oliverneumann usingweatherdatainenergytimeseriesforecastingthebenefitofinputdatatransformations
AT marianturowski usingweatherdatainenergytimeseriesforecastingthebenefitofinputdatatransformations
AT ralfmikut usingweatherdatainenergytimeseriesforecastingthebenefitofinputdatatransformations
AT veithagenmeyer usingweatherdatainenergytimeseriesforecastingthebenefitofinputdatatransformations
AT nicoleludwig usingweatherdatainenergytimeseriesforecastingthebenefitofinputdatatransformations