Por favor, use este identificador para citar o enlazar este ítem: https://hdl.handle.net/11000/30981

Comparison of Short-Term Load Forecasting Performance by Neural Network and Autoregressive Based Models

Título :
Comparison of Short-Term Load Forecasting Performance by Neural Network and Autoregressive Based Models
Autor :
López, Miguel
VALERO, SERGIO  
Senabre, Carolina
Gabaldón, Antonio
Editor :
Institute of Electrical and Electronics Engineers
Departamento:
Departamentos de la UMH::Ingeniería Mecánica y Energía
Fecha de publicación:
2018-06-27
URI :
https://hdl.handle.net/11000/30981
Resumen :
In the past decade, many techniques ranging from statistical methods to complex artificial intelligence systems have been proposed by implementing their application to an electric system and highlighting its performance; usually providing a measure of accuracy like RMSE over a definite period. However, there is little research in which a fair comparison among methods is demonstrated, and it is difficult to determine which method would be better suited to a particular electric system or data set. This paper analysis one of the forecasting models running at the National Transport Operator of the Spanish system (REE), which is based on both autoregressive and neural network techniques. The results of this paper help to determine under which circumstances each of the models shows a better performance, which periods are more accurately forecasted by each model and provide valid criteria to choose one or the other depending on the characteristics of the application.
Palabras clave/Materias:
autoregressive processes
demand forecasting
neural networks
power demand
Área de conocimiento :
CDU: Ciencias aplicadas: Ingeniería. Tecnología
Tipo documento :
application/pdf
Derechos de acceso:
info:eu-repo/semantics/openAccess
Attribution-NonCommercial-NoDerivatives 4.0 Internacional
DOI :
https://doi.org/10.1109/EEM.2018.8469797
Aparece en las colecciones:
Ponencias y Comunicaciones Ingeniería Mecánica y Energía



Creative Commons La licencia se describe como: Atribución-NonComercial-NoDerivada 4.0 Internacional.