Modélisation de la consommation électrique à partir de grandes masses de données pour la simulation des alternatives énergétiques du futur

Abstract : Future trend of electricity demand is a key point for sizing both the electricity network and the power plants. In order to forecast future electricity demand, current models mostly use statistical approaches based on past demand measurements and on demographic and economic trends. Because of current context of energy transition which comes along with important changes, future electricity demand is not expected to be similar to past trends. Modeling these changes requires a bottom-up modeling of each contributor to electricity demand. This kind of model is challenging because of the large number of input data required. At the same time, data and information are more and more available. Such availability can be considered both as an asset for modeling and as an important issue because of data heterogeneity. In this context, this dissertation offers an approach to build a bottom-up load curve simulator which enables to simulate prospective scenarii at the scale of France country. Firstly, an assessment, classification, and matching of the large databases explaining the electricity demand have been performed. Then, the electricity demand model has been presented. It has been validated and calibrated on Enedis’ large volumes of electricity demand measurements of medium voltage feeders. Finally, this model has been used to simulate several prospective scenarii in order to improve the electricity distribution network sizing.
Document type :
Theses
Complete list of metadatas

Cited literature [128 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/tel-01774316
Contributor : Abes Star <>
Submitted on : Monday, April 23, 2018 - 3:19:06 PM
Last modification on : Wednesday, May 15, 2019 - 12:37:06 PM
Long-term archiving on: Tuesday, September 18, 2018 - 11:44:26 PM

File

2017PSLEM032_archivage.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-01774316, version 1

Citation

Thibaut Barbier. Modélisation de la consommation électrique à partir de grandes masses de données pour la simulation des alternatives énergétiques du futur. Energie électrique. PSL Research University, 2017. Français. ⟨NNT : 2017PSLEM032⟩. ⟨tel-01774316⟩

Share

Metrics

Record views

633

Files downloads

3916