A crack localization method for beam-analogy structures based on wavelet transform of mode shape variations
This study proposes a crack localization method for beam-analogy structures with and without inherent structural discontinuities such as girders and bulkheads of ships, bridge piers, and building slabs. The proposed method utilizes the discrete wavelet transform (DWT) of crack-induced mode shape variations to identify the structural singularity due to a crack regardless of the singularity of mode shape curvature caused by the inherent structural discontinuity. In addition, multiple mode shapes with different nodal and mode shape curvature distributions as well as a polynomial padding mitigating edge effects are utilized to enhance crack detectability. The Reconstructed Detail Component (RDC) derived from the inverse wavelet transform of detail coefficients is used as a crack indicator to preserve spatial resolution. The validity of the proposed method is examined through numerical analyses on a uniform beam, a ship-like box-girder structure with bulkheads, and a real container ship structure. The results confirm that the proposed method is able to localize a crack and to estimate its severity, even when the crack occurs at an inherent structural discontinuity. • Crack localization method for beam-analogy structures based on wavelet transform are proposed. • Mode shape variations are utilized to identify a crack-induced singularity of the mode shape curvature. • Reconstructed Detail Component (RDC) is used as a crack diagnosis indicator to preserve spatial resolution of mode shape. • Validity and applicability of the proposed method are investigated through finite element analysis. • Proposed method successfully localizes a crack and estimates its severity, even when it occurs at an structural discontinuity.
Publication Details
- Journal
- International Journal of Naval Architecture and Ocean Engineering
- Published
- 2025-12-12
- DOI
- https://doi.org/10.1016/j.ijnaoe.2025.100726
- Primary Topic
- Structural Health Monitoring Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00