Title: Locating damages in beams with artificial neural network

Authors: Marília Marcy; Andrea Brasiliano; Gustavo Barbosa Lima Da Silva; Graciela Doz

Addresses: Department of Civil and Environmental Engineering, University of Brasília, UnB-Campus Universitário Darcy Ribeiro, Asa Norte, 70910-900 – Brasilia, DF, Brazil ' Federal University of Paraíba, Department of Civil and Environmental Engineering, UFPB – Centro de Tecnologia – Campus I, Cidade Universitária – Joao Pessoa, 58000-000, PB, Brazil ' Federal University of Paraíba, Department of Civil and Environmental Engineering, UFPB – Centro de Tecnologia – Campus I, Cidade Universitária – Joao Pessoa, 58000-000, PB, Brazil ' Department of Civil and Environmental Engineering, University of Brasília, UnB-Campus Universitário Darcy Ribeiro, Asa Norte, 70910-900 – Brasilia, DF, Brazil

Abstract: Despite the significant advance of new technologies applied in civil engineering, the structures continue subjected to the occurrence of faults which are produced in normal conditions of use. The detection of these faults may be done by the analysis of some dynamic characteristics as frequencies and mode shapes. This fact implies that such characteristics must be identified accurately in order to produce reliable results about structural health. Artificial neural network may be also used as an important tool in the evaluation of structural integrity. This work presents an experimental analysis, in which a laboratory beam was submitted to different vibration tests. In the attempt to establish a suitable methodology for structural evaluation, the method of artificial neural network was applied in order to identify the damage location and the obtained results were very satisfactory.

Keywords: artificial neural networks; ANNs; dynamic properties; damage identification; damage detection; damaged beams; structural integrity; vibration tests; structural evaluation; damage location.

DOI: 10.1504/IJLCPE.2014.064110

International Journal of Lifecycle Performance Engineering, 2014 Vol.1 No.4, pp.398 - 413

Received: 04 Oct 2013
Accepted: 06 May 2014
Published online: 30 Aug 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article