Skip to Main content Skip to Navigation
Master Thesis

La notion d’année typique en météorologie

Abstract : The typical year problem comes from a need of meteorological data users to reduce the datasets’ size because of the computational cost of their models. A state of art allowed identifying two key steps in the obtention of typical years, and changings have been suggested and tested at both levels in order to identify the questions to be asked and the methodologies to be used when it comes to obtaining typical years. The first step consists in building the candidates to the typical year election. Instead of an extraction of the real years, a generator by blocks of classes has been suggested. The second step is about chosing the most typical candidate. Several measures of dissimilarity for time series have been suggested and studied. The different methodologies have been tested in two projects: the first one is DESIRES, supported by ERANET/MED, it aims at dimensioning a desalination system supplied by wind generators and photovoltaics. The second one is MEDISA by Eau du Ponant, which aims at dimensioning a retention basin for rainfalls.
Document type :
Master Thesis
Complete list of metadatas

Cited literature [4 references]  Display  Hide  Download
Contributor : Agrocampus Ouest <>
Submitted on : Thursday, November 9, 2017 - 10:03:37 AM
Last modification on : Monday, July 6, 2020 - 3:38:22 PM


Files produced by the author(s)


  • HAL Id : dumas-01631365, version 1



Marie Boutigny. La notion d’année typique en météorologie. Statistiques [stat]. 2017. ⟨dumas-01631365⟩



Record views


Files downloads