Skip to Main content Skip to Navigation
Master Thesis

Nouvelles fonctionnalités, évolutions et optimisations du logiciel l'assimilation variationnelle Yao

Abstract : The Yao software is a source code generator which aims to perform variational assimilation runs. One of its fonctionnalities is to help the user to conceive numeric simulations by using the notion of a modular graph for model representation. It gives an integrated development environment which provides the user with the ability to make standard and incremental variational assimilation sessions. Furthermore, it supplies a number of test and control functions. In this dissertation, we present the evolutions developed in the Yao software during the internship of Guillaume Rosinosky. A refactoring of the source code of the generated project has been done in order to simplify the maintenance and the readability of the source code, by using oriented object representations of the internal concepts of Yao. Two new algorithms have been incorporated and tested. The implementation of the quasi-static algorithm open the possibility to assimilate more complex models. The variational assimilation in weak constraint algorithm allows to take into consideration the model errors by switching in dual form the variational assimilation equations. In addition to those major modifications, general optimisation of the performances have been applied.
Document type :
Master Thesis
Complete list of metadata

Cited literature [16 references]  Display  Hide  Download
Contributor : Cnam - Service Commun De La Documentation - Bibliothèque Centrale Connect in order to contact the contributor
Submitted on : Tuesday, April 29, 2014 - 2:03:49 PM
Last modification on : Friday, August 5, 2022 - 2:54:00 PM
Long-term archiving on: : Tuesday, July 29, 2014 - 12:25:12 PM


  • HAL Id : dumas-00985237, version 1



Guillaume Rosinosky. Nouvelles fonctionnalités, évolutions et optimisations du logiciel l'assimilation variationnelle Yao. Génie logiciel [cs.SE]. 2012. ⟨dumas-00985237⟩



Record views


Files downloads