Presentation
Concept Drift Detection to Improve Time Series Forecasting of Wind Energy Generation
Author/s | Cabello López, Tomás
Cañizares Juan, Manuel Carranza García, Manuel García Gutiérrez, Jorge Riquelme Santos, José Cristóbal |
Department | Universidad de Sevilla. Departamento de Lenguajes y Sistemas Informáticos |
Publication Date | 2022 |
Deposit Date | 2022-12-12 |
ISBN/ISSN | 978-3-031-15470-6 0302-9743 |
Abstract | Most of the current data sources generate large amounts of
data over time. Renewable energy generation is one example of such data
sources. Machine learning is often applied to forecast time series. Since
data flows are ... Most of the current data sources generate large amounts of data over time. Renewable energy generation is one example of such data sources. Machine learning is often applied to forecast time series. Since data flows are usually large, trends in data may change and learned pat terns might not be optimal in the most recent data. In this paper, we analyse wind energy generation data extracted from the Sistema de Infor mación del Operador del Sistema (ESIOS) of the Spanish power grid. We perform a study to evaluate detecting concept drifts to retrain models and thus improve the quality of forecasting. To this end, we compare the performance of a linear regression model when it is retrained randomly and when a concept drift is detected, respectively. Our experiments show that a concept drift approach improves forecasting between a 7.88% and a 33.97% depending on the concept drift technique applied |
Funding agencies | Ministerio de Ciencia e Innovación (MICIN). España Junta de Andalucía |
Project ID. | PID2020-117954RB-C22
US-1263341 P18-RT-2778 |
Files | Size | Format | View | Description |
---|---|---|---|---|
978-3-031-15471-3_12.pdf | 1.151Mb | [PDF] | View/ | |