Adaptive detection of structural changes based on unsupervised learning and moving time-windows
SANTOS ; CALADO ; ORCESI ; CREMONA
Type de document
COMMUNICATION AVEC ACTES INTERNATIONAL (ACTI)
Langue
anglais
Auteur
SANTOS ; CALADO ; ORCESI ; CREMONA
Résumé / Abstract
The present paper addresses data-driven structural health monitoring to propose a real-time strategy for adaptive structural assessment. The adaptive character is achieved using unsupervised discrimination machine-learning methods, widely known as clustering algorithms. Real-time capability is based on the definition of symbolic data, which allow describing large amounts of information without loss of related information. The efficiency of the proposed methodology is illustrated using an experimental case study in which structural changes were imposed to a suspended bridge during an extensive rehabilitation program.
