Abstract:

To address the challenge of efficiently identifying internal voids and compactness defects in CFST structures, a non-contact vibration detection and compactness screening method based on a laser Doppler vibrometer was proposed. A Gaussian curvature index of the flexibility matrix was constructed to characterize the curvature variation of surface vibration responses, while a Bayesian fusion algorithm was introduced to achieve adaptive integration of multi-source vibration data. On this basis, an on-site applicable framework for compactness identification and evaluation was established. The results show that the finite element model is in good agreement with the experimental results in the first four modal parameters, demonstrating that the model reasonably represents the dynamic characteristics of CFST structures. Field tests further confirm that the proposed approach can extract dominant modal parameters under complex construction noise and identify both shallow and deep defects with high accuracy and robustness.

Key words: concrete-filled steel tubular (CFST), laser Doppler, compactness, least squares complex exponential method (LSCE), engineering application