Aller au contenu  Aller au menu  Aller à la recherche

Bienvenue - Laboratoire Jacques-Louis Lions




Print this page |


Chiffres clefs

217 personnes travaillent au LJLL

83 personnels permanents

47 enseignants chercheurs

13 chercheurs CNRS

9 chercheurs INRIA

2 chercheurs CEREMA

12 ingénieurs, techniciens et personnels administratifs

134 personnels non permanents

85 doctorants

16 post-doc et ATER

5 chaires et délégations

12 émérites et collaborateurs bénévoles

16 visiteurs


Chiffres janvier 2014


Séminaire du LJLL : R. DeVore

22 mai 2015 — 14h00
Ron DeVore (Université A&M du Texas)
Data assimilation in solving parametric PDEs
 This talk is concerned with the following problem. We wish to recover the solution u(a*) to a known parametric family of PDEs at a certain parameter value a* that is unknown to us. However, we have information about the state u(a*) through some set of physical measurements which can be viewed as the application of linear functionals to u(a*). How should we merge these two pieces of information, the parametric model and the measurements, to effectively recover u(a*) ?
 The parametric model is complex and the solution manifold is usually known only through a sequence of known finite dimensional spaces V_0, … ,V_n with dim(V_k) = k that are known to approximate the solution manifold to a known accuracy epsilon_k. We formulate this as an optimal recovery problem and determine the optimal solution. Our results clarify and extend the fundamental work of Maday, Patera, Penn and Yano.