Partenaires

CNRS
EDF
CEA


Rechercher

Sur ce site

Sur le Web du CNRS


Accueil du site > Research Areas

Data identification, assimilation, exploitation, reduction and coupled problems involving structures

par Alexandre MARTIN - 7 août 2013

Control of the mechanical integrity of large industrial structures during their lifetime needs to investigate in different fields than the one used for conception or dimensioning. An important part of this research operation relies on the feedback from real operated structures and questions linked to their exploitation, and the development of in service monitoring methodologies as well as predictive tools for their behaviour in the future.

We will investigate more specifically :

  • Inverse problems : with a specific focus on reciprocity gap methods and questions concerning the prolongation of surface data which has today numerous applications exploiting the development of field measure techniques ;
  • Exploration of data assimilation processes, initially developed for meteorological or oceanographic applications, in the context of non linear mechanics, with the objective of long term estimation of the evolution of nuclear reinforced concrete buildings ;
  • Analyses of coupled phenomena which were not taken into account during the conception phase of the design of pipe networks and which appeared during exploitation : fluid-structure interaction with instability analysis leading to cross-flow hissing at diaphragms ;
  • Prediction of the effect of a structural modification on its vibratory in service behaviour in order to eliminate unwanted resonance. The chosen approach is hybrid : it exploits on site experimental data and a numerical representation of the structural modification only ;

  • A last, realistic models for cracked rotors (representation of 3D cracks with contact), in order to be able to identify which indicators are the most pertinent to design new in service surveillance monitoring of these structures.