Damage Localization Based on MUSIC Beamforming Algorithm Under Variable Temperature Conditions

Abstract

Environmental temperature variations pose significant challenges in Lamb wavebased structural health monitoring by introducing amplitude and phase distortions that can mask damage-related signal features. This paper presents a novel damage imaging and localization method integrating an improved Multiple Signal Classification (MUSIC) algorithm with beamforming techniques to address these challenges. An array signal propagation model has been developed to explicitly account for temperatureinduced effects on wave propagation characteristics, while an enhanced MUSIC algorithm incorporating amplitude-phase error compensation through cost function optimization has been formulated to mitigate signal distortions. The method further employs beamforming-based spatial filtering to improve imaging resolution under varying temperature conditions. The effectiveness of the proposed method has been validated through extensive experimental studies conducted on aluminum plates with square hole damages under temperatures ranging from -40°C to 80°C. Results demonstrate that the method achieves consistent damage localization accuracy with relative errors maintained below 5.69% across all temperature conditions, while successfully preserving imaging resolution and boundary definition. The method's robust performance and computational efficiency make it particularly suitable for practical structural health monitoring applications where temperature variations are inevitable.