Abstract Background Concrete-Filled Steel Tube (CFST) composite columns are widely utilized in civil engineering structures due to their high structural efficiency and superior load-bearing capacity. However, interfacial debonding between the steel tube and the concrete core constitutes a critical damage mechanism that can significantly alter the dynamic response and compromise structural integrity. Purpose This study proposes a novel vibration-driven computational methodology for detecting and localizing interfacial debonding in CFST columns. Methods The proposed methodology employs curvature analysis of the initial (intact) and current vibration mode shapes and introduces the Irregularity Detection Index (IDI) as a robust damage indicator. Vibration mode shapes are obtained through finite element analysis and are subsequently post-processed within a numerical computing platform to evaluate the IDI and identify damage locations. Results Numerical results demonstrate that the proposed algorithm accurately detects interfacial debonding between the concrete core and the steel tube. The method exhibits high sensitivity in detecting damage near boundary edges and at the column base, regions that are typically challenging for structural health monitoring (SHM). Furthermore, reliable diagnostic performance is maintained under low-level simulated noise, confirming the robustness of the approach. Conclusions The proposed vibration-based computational methodology provides an accurate and robust tool for detecting and localizing interfacial debonding in CFST composite columns. Its effectiveness in identifying damage in boundary regions and at the column base, together with its stability under low-level noise conditions, highlights its strong potential for practical SHM applications.
Tarbozagh et al. (2026) studied this question.