부산대학교 도서관

Low excitation voltage (LEV) electromagnetic ultrasonic detection is needed for the safety of high-temperature boilers, reactors, pipelines, and other equipment. However, the low signal intensity and high noise due to the LEV and the high-temperature environment result in a low signal-to-noise ratio (SNR) in the defect detection process. This study proposes a novel noise reduction method, called VMD-singular value decomposition (SVD)-CEEMD (VSCE), which utilizes variational mode decomposition (VMD), SVD, and complementary ensemble empirical mode decomposition (CEEMD) to extract defect information and suppress pulsating noise, instrument noise, and high-temperature structural noise. The shear wave generated by an electromagnetic acoustic transducer was used to detect two different materials at 25 V LEV, with a temperature range of 25 °C–700 °C. The VSCE method was found to improve the SNR of received signals by two to three times compared to conventional denoising methods, such as wavelet threshold denoising (WTD), VMD, and empirical mode decomposition (EMD). The positioning error of the defect location is less than 2.06%, which has proven the great application prospects and value of the VSCE methods in the LEV detection of high-temperature equipment.