Use of symbolic data analysis for structural health monitoring applications - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2010

Use of symbolic data analysis for structural health monitoring applications

Filipe Afonso
  • Fonction : Auteur
Edwin Diday
  • Fonction : Auteur
  • PersonId : 858027
Norbert Badez
  • Fonction : Auteur
Yves Genest
  • Fonction : Auteur
Magali Claudel
  • Fonction : Auteur

Résumé

Structural health monitoring (SHM) provides accurate in situ information and its use in management of infrastructures is an important challenge. For a large population of assets, SHM information may be composed of numerous and various types of data. The crucial issue is that the quantity and heterogeneity of information may finally prevent from efficiently correlating collected data and informing the performance metrics of interest. Data mining concepts offer possibilities of analyzing and correlating complex (i.e., heterogeneous) data. The fusion of these data to get new knowledge requires using symbolic data which are an extension of standard numerical or categorical data. Symbolic Data Analysis (SDA) is illustrated in this paper by the study of the degradation problems occurring on nuclear power plant cooling towers. This paper presents the suitability of SDA for structural degradation analysis, combination of heterogeneous measures, description of the correlations between them, and classification of the degradations. Complex data
Fichier non déposé

Dates et versions

hal-01066316 , version 1 (19-09-2014)

Identifiants

  • HAL Id : hal-01066316 , version 1

Citer

Filipe Afonso, Edwin Diday, Norbert Badez, Yves Genest, Magali Claudel, et al.. Use of symbolic data analysis for structural health monitoring applications. IALCCE 2010 - International Symposium of Life-Cycle Civil Engineering, Oct 2010, France. pp 205-210. ⟨hal-01066316⟩
77 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More