Skip to Main content Skip to Navigation
Theses

Assimilation de données variationnelle pour les problèmes de transport des sédiments en rivière

Abstract : The prediction of the sedimentation in rivers requires the utilization of mathematical models governing the flow and observed data. The objective of this work is to propose a data assimilation method to retrieve physical fields by combining the model and the observation. This method is based on optimal control techniques. We present sedimentation problems and their numerical approximations, as well as a splitting algorithm and a study on convergence is carried out. Before working on reel problems, variational data assimilation methods are developed and tested its feasibility for three types of sediment transport problems : 1) determination of initial condition, 2) identification of parameters, 3) estimation of modeling errors. As the studies on real sedimentation fields will lead to numerical problems of great dimensionality, we have been interested, in the last part, in the techniques of reducing the size of the control space in order to obtain problems with reasonable size.
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-00004863
Contributor : Thèses Imag <>
Submitted on : Wednesday, February 18, 2004 - 6:28:17 PM
Last modification on : Friday, November 6, 2020 - 4:04:35 AM
Long-term archiving on: : Friday, April 2, 2010 - 7:36:16 PM

Identifiers

  • HAL Id : tel-00004863, version 1

Collections

Citation

Junqing Yang. Assimilation de données variationnelle pour les problèmes de transport des sédiments en rivière. Modélisation et simulation. Université Joseph-Fourier - Grenoble I, 1999. Français. ⟨tel-00004863⟩

Share

Metrics

Record views

450

Files downloads

202