Spectral algorithm for non-destructive damage localisation: Application to an ancient masonry arch model. (1st February 2017)

• Record Type: Journal Article
• Authors: Masciotta, Maria-Giovanna; Ramos, Luís F.; Lourenço, Paulo B.; Vasta, Marcello
• Is Part Of: Mechanical systems and signal processing. Volume 84:Part A(2017)
• Page Start: 286
• Page End: 307
• Publication Date: 2017-02-01
• Subjects: Non-destructive damage localisation -- Eigenvalue decomposition -- Spectral-based method -- Historical structures -- Masonry arch; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2016.06.034
• Languages: English
• ISSNs: 0888-3270

Abstract

Abstract: Structural monitoring and vibration-based damage identification methods are fundamental tools for condition assessment and early-stage damage identification, especially when dealing with the conservation of historical constructions and the maintenance of strategic civil structures. However, although the substantial advances in the field, several issues must still be addressed to broaden the application range of such tools and to assert their reliability. This study deals with the experimental validation of a novel method for non-destructive damage identification purposes. This method is based on the use of spectral output signals and has been recently validated by the authors through a numerical simulation. After a brief insight into the basic principles of the proposed approach, the spectral-based technique is applied to identify the experimental damage induced on a masonry arch through statically increasing loading. Once the direct and cross spectral density functions of the nodal response processes are estimated, the system's output power spectrum matrix is built and decomposed in eigenvalues and eigenvectors. The present study points out how the extracted spectral eigenparameters contribute to the damage analysis allowing to detect the occurrence of damage and to locate the target points where the cracks appear during the experimental tests. The sensitivity of the spectral formulation to the level of noise in the modal data is investigated and discussed. As a final evaluation criterion, the results from the spectrum-driven method are compared with the ones obtained from existing non-model based damage identification methods. Highlights: A spectral-based damage identification algorithm is validated experimentally. Multiple damage scenarios are statically induced on a masonry arch replica. The spectral properties of the system are analysed over different damage scenarios. Cracks are detected and located through a spectral-based damage index. The sensitivity of the spectral formulation to noisy modal data is investigated.