Abstract
This paper proposes a methodology for developing multi-site damage location systems for engineering structures that can be trained using single-site damaged state data only. The methodology involves training a sequence of binary classifiers based upon single-site damage data and combining the developed classifiers into a robust multi-class damage locator. In this way, the multi-site damage identification problem may be decomposed into a sequence of binary decisions. In this paper Support Vector Classifiers are adopted as the means of making these binary decisions. The proposed methodology represents an advancement on the state of the art in the field of multi-site damage identification which require either: (1) full damaged state data from single- and multi-site damage cases or (2) the development of a physics-based model to make multi-site model predictions. The potential benefit of the proposed methodology is that a significantly reduced number of recorded damage states may be required in order to train a multi-site damage locator without recourse to physics-based model predictions. In this paper it is first demonstrated that Support Vector Classification represents an appropriate approach to the multi-site damage location problem, with methods for combining binary classifiers discussed. Next, the proposed methodology is demonstrated and evaluated through application to a real engineering structure – a Piper Tomahawk trainer aircraft wing – with its performance compared to classifiers trained using the full damaged-state dataset.
Highlights
This study presents a promising approach for handling multi-site damage location problems without requiring training data from all damage states
It is shown that a multi-class classifier based on binary support vector machines (SVMs) is a suitable approach for handling multi-class damage identification problems: for the experimental case presented, near perfect results were achieved when damaged-state data for all the structural states are incorporated in the training set
The second, and major, contribution of the study is to show that a classifier trained using only normal condition and single-site damage data may be capable of identifying the presence of multi-site damage with a high degree of accuracy
Summary
Multi-site damage identification represents an important and challenging problem in SHM but one that has received comparatively little dedicated attention in the literature, with the majority of approaches presented in the literature (see, for example, the extensive reviews in [3,4]) focusing on the identification of single-site damage. The impact of this restriction is clear given that an in situ structure would be expected, over time, to exhibit degradation from its baseline state concurrently at multiple locations. Such features exhibit disadvantages alongside advantages (susceptibly to experimental noise being an example in the case of modeshape curvatures) and there is clear motivation for developing methods for multi-site damage location that maintain generality of feature choice
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
