| Abstract [eng] |
This dissertation advances Guided Wave (GW)-based Structural Health Monitoring (SHM) by addressing critical challenges in damage detection for large metallic/composite structures (e.g., aircraft, wind turbines). It introduces three methodologies to enhance GW signal processing and imaging: (1) A streamlined framework reduces reliance on user-defined parameters in GW analysis, prioritizing versatile, low-variable techniques to improve adaptability across materials and defect types. (2) For RAPID imaging—a core SHM method—a novel post-processing strategy suppresses interference artifacts caused by overlapping probability maps without altering the core algorithm or introducing new variables, validated on a carbon fiber-reinforced plastic laminate plate with simulated localized defects. (3) A baseline-free approach for localized damage imaging using filtered back projection overcomes the absence of baseline signals via mean-based sinogram thresholding, reconstructing defect geometry in glass fiber-reinforced plastic with impact damage. By integrating robust signal conditioning, feature extraction, and imaging innovations, this research bridges theoretical and practical gaps, delivering computationally efficient, field-deployable solutions for high-resolution damage mapping and quantification in critical infrastructure, enhancing reliability for real-world SHM applications. |
