Lamb Wave Damage Localization under Variable Temperature Using Dual Array MUSIC Beamforming with Joint Amplitude and Phase Compensation

Abstract : Temperature variations can significantly alter Lamb-wave amplitude, phase, and propagation velocity, thereby degrading damage localization and imaging performance in structural health monitoring. This study proposes a temperature-robust dual-array MUSIC–beamforming method for Lamb-wave damage localization. A temperature-dependent near-field propagation model is established to represent the combined amplitude and phase distortions using an element-wise complex error matrix. A cost-function optimization scheme is then developed to jointly estimate the damage location and temperature-induced compensation parameters through alternating updates, without requiring a complete reference signal at every temperature. The compensated MUSIC spectrum is further fused with a delay-and-sum beamforming map to combine high angular resolution with physically meaningful signal-energy distribution. Experiments were conducted on three Al6061 plates containing artificial defects over a temperature range of −40 to 80 °C. At five calibrated temperatures, the relative localization errors ranged from 1.76% to 8.12%. At six intermediate temperatures without direct calibration, the maximum error was 9.52%, with a mean error of 6.99%. The method also maintained gradual performance degradation under additive white Gaussian noise, with the maximum error reaching 14.1% at 0 dB SNR. These results demonstrate that the proposed method provides accurate and robust Lamb-wave damage localization and improved image contrast under variable-temperature conditions.