摘要

To solve the problem of low signal-to-noise ratio and resolution caused by the non-stationary signals in the defect detection of thick-walled pipelines with the circumferential piezoelectric Lamb waves, the phase coherence imaging algorithm (PCI) is used to extract the phase information from the Lamb wave signals to construct the phase coherence factor, and the dynamic weighting process is used to amplify the contribution of the phase distribution to the pixel amplitude of the ultrasound images. On the one hand, the method preserves the defect echo amplitude with the same phase distribution and enhances the signal to noise ratio. On the other hand, this method can improve the virtual focusing effect and lateral resolution of the post-processed image by enhancing the directivity of the aperture beam. The flaw detection of a thick-walled pipeline with an outer diameter of ?269 mm and wall thickness of 32 mm with a transverse crack of 20 mm has been carried out, and the obtained signals have been reconstructed. The results show that the signal to noise ratio of PCI is 25.2 dB, which is 15.6 dB higher than that of the original B scan images, and the half-wave high horizontal width is reduced from 47.3 mm to 22.4 mm, which improves the horizontal resolution and improves the recognition degree of defects.