Descripción
|
|
---|---|
The management of an electrical system requires knowing the energy demand well in advance. A common practice is to model the hourly time series of electricity demand. The strong daily seasonality has led many authors to estimate a different model for each hour of the day. This article presents a new approach that assembles the 24 hourly series in a periodic autoregressive moving-average model. The identification and estimation of a periodic model of order 24 presents enormous complexity. In this paper we present a theoretical result that greatly simplifies this task. The paper presents an original method of estimating the periodic model that takes advantage of the existing implementation of estimating univariate ARIMA models and describes their application in the prediction of the next hours. The new methodology includes two additional contributions: (1) a very exhaustive and complex intervention system that allows the reduction of prediction errors occurring during non-working days, and (2) a meticulous model of the non-linear temperature effect using regression spline techniques. The method is currently being used by the Spanish System Operator (\emph{Red El\'{e}ctrica de Espa\~{n}a}, REE) to make hourly forecasts of electricity demand from one to ten days ahead. | |
Internacional
|
Si |
Nombre congreso
|
20th IIF Workshop | Predictive Energy Analytics in the Big Data World Cairns, Australia 22-23 June 2017 |
Tipo de participación
|
960 |
Lugar del congreso
|
Cairns, Australia |
Revisores
|
Si |
ISBN o ISSN
|
0000-0000 |
DOI
|
|
Fecha inicio congreso
|
22/06/2017 |
Fecha fin congreso
|
23/06/2017 |
Desde la página
|
1 |
Hasta la página
|
37 |
Título de las actas
|
- |