A two-stage guided wave-based framework for damage localization and shape approximation using a probabilistic binary topology optimization algorithm
Date
2025
Authors
Zeng, Z.
Feng, Y.
Ng, C.T.
Sheikh, A.H.
Editors
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Journal article
Citation
Engineering structures, 2025; 343(C):121149-1-121149-20
Statement of Responsibility
Zijie Zeng, Yuan Feng, Ching Tai Ng, Abdul Hamid Sheikh
Conference Name
Abstract
Guided wave (GW) model-based methods have demonstrated strong capabilities in damage characterization by calibrating damage parameters to minimize discrepancies between simulations and measurements. Unlike baseline-dependent approaches, GW model-based methods rely on physics-based or data-driven models, thereby eliminating the need for experimentally measured reference datasets. This makes them more suitable for complex damage scenarios. However, most existing GW model-based methods rely on prior knowledge or assumptions about damage types, which limits their practicality. This paper presents a two-stage GW model-based framework for damage localization and shape approximation. GW propagation is simulated using a higher-order finite element model, while damage is represented by independent void elements, allowing for the simulation of various damage geometries. In Stage One, a damage index scheme is used to identify the potential damage region. In Stage Two, an inverse binary optimization problem is formulated and solved to determine the damage state (i.e., intact or damaged) of each element within the identified region, following a concept similar to topology optimization. The performance of the proposed framework is evaluated for identifying different surface cracks in beam-like structures. A probabilistic discrete binary harmony search algorithm with Tabu search and enhanced local search method (DHBS-TS-ELSM) is proposed to solve high-dimensional binary optimization problems. Numerical case studies and experimental tests were carried out to assess the effectiveness of the proposed framework in identifying different types of surface cracks. The application of topology-based concepts for guided wave-based damage identification demonstrates strong potential and offers a practical direction for further research and development.
School/Discipline
Dissertation Note
Provenance
Description
Access Status
Rights
© 2025 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).