Research on the application of F-P acoustic detection technology for subsurface micro-defect identification

Precise detection of sub-millimeter micro-defects in materials remains a significant challenge due to their weak scattering signals and the non-stationary characteristics they induce, which are often easily masked by noise. The proposed Fabry-Perot (F-P) acoustic detection technology is a non-contact and non-destructive method that has shown high sensitivity in detecting sub-millimeter micro-defects during preliminary experimental verification. In this study, an F-P acoustic detection system was developed with enhanced detection capability achieved by reducing the laser spot size. Validation experiments were performed on aluminum alloy specimens with artificial defects. An innovative adaptive weighted wavelet singular spectral entropy (AW-WSSE) feature extraction algorithm was proposed, which integrates wavelet decomposition, singular spectral entropy (SSE), and adaptive weighted fusion. Compared with conventional time/frequency-domain methods and traditional SSE, the proposed method significantly improves imaging quality and successfully detects micro-defects as small as 0.2 mm. Under the conditions of this study, compared to phased array ultrasonic testing (PAUT), the spatial resolution is increased by at least 2.875 times, the signal-to-noise ratio (SNR) is approximately 2.231 times higher, and relative error reduced by 34.4 %. These results demonstrate the superior detection performance and potential of the proposed method.