Adaptive damage localization based on locally perturbed dynamic equilibrium and hierarchical clustering

Cao, Shancheng, Ouyang, Huajiang ORCID: 0000-0003-0312-0326 and Cheng, Li (2019) Adaptive damage localization based on locally perturbed dynamic equilibrium and hierarchical clustering. SMART MATERIALS AND STRUCTURES, 28 (7). 075003-075003.

Text SMS-107893R1.docx - Author Accepted Manuscript Download (9MB)

Abstract

Pseudo-excitation (PE) method is a recently developed damage identification method for flexible structures containing components like beams, plates and shells. Characterized by the high-order spatial derivatives, the approach has been shown to feature a high sensitivity to local damage. However, two major issues, i.e. susceptibility to measurement noise and unknown material/structural properties, hamper its practical applications. To tackle these problems, an adaptive damage localization method is proposed for plate-type structures, which combines the PE method with hierarchical clustering. In the proposed method, a general dynamic equilibrium model, involving unknown material/structural properties, is statistically identified and further used for damage localization. Moreover, noise-induced effects are quantified by using a hierarchical clustering for performance assessment of damage localization and process optimization of spatial derivative estimation to achieve more accurate damage localization. Meanwhile, a data fusion scheme is developed to avoid blind inspection zones, thus enhancing the capability of damage localization. Both numerical and experimental studies of cantilever plates containing two damage zones are conducted to validate the feasibility and the effectiveness of the proposed adaptive damage localization method. Results demonstrate that the proposed method outperforms the traditional PE method in terms of detection accuracy and robustness.

Item Type:	Article
Uncontrolled Keywords:	damage localization, local dynamic equilibrium, pseudo-excitation, spatial derivatives, hierarchical clustering
Depositing User:	Symplectic Admin
Date Deposited:	19 Jun 2019 14:39
Last Modified:	19 Jan 2023 00:39
DOI:	10.1088/1361-665X/ab1abe
Related URLs:	Author Publisher
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3046554