Eventos Anais de eventos
COBEM 2019
25th International Congress of Mechanical Engineering
ART-2 NEURAL NETWORK FOR DAMAGE CLASSIFICATION IN IMPEDANCE-BASED SHM
Submission Author:
Stanley Washington Ferreira Rezende , GO
Co-Authors:
Stanley Washington Ferreira Rezende, Bruno Pereira Barella, Renan Garcia Rosa, Robson Medrado de Oliveira, Jose dos Reis Vieira de Moura Jr
Presenter: Stanley Washington Ferreira Rezende
doi://10.26678/ABCM.COBEM2019.COB2019-2459
Abstract
This work aims to present the use of the ART-2 Neural Network in the intelligent diagnosis of structural failures. For this, the technique of impedance-based structural health monitoring is used in order to inspect a hydraulic pipeline composed of PVC pipes. A simulated damage by mass addition was inserted near the influence range of the PZT patch, and subsequently data processing techniques as temperature compensation were applied to impedance signatures (baseline and damage) in order to identify possible misdiagnosis. After applying the ART-2 Neural Network, it can be observed that it was able to efficiently delineate the different states of pipeline integrity, being applicable to the monitoring of structural integrity.
Keywords
Impedance-based Structural Health Monitoring, ART-2 Neural Network, Smart Diagnostics
