Fractional Fourier Transform-Based Signal Separation for Ultrasonic Guided Wave Inspection of Plates.

Detecting defects in plates is crucial across various industries due to safety risks. While ultrasonic bulk waves offer point-by-point inspections, they are time-consuming and limited in coverage. In contrast, guided waves enable the rapid inspection of larger areas. Array transducers are typically used for more efficient coverage, but conventional excitation methods require sufficient time delays between the excitation of array elements that prolong inspection time, necessitating data acquisition time optimization. Reducing time delays can lead to signal overlapping, complicating signal separation. Conventional frequency domain or time-domain filtering methods often yield unsatisfactory separation results due to the signal overlapping in both domains. This study focuses on the application of the Fractional Fourier Transform (FrFT) for separating overlapping ultrasonic signals, leveraging the FrFT's ability to distinguish signals that overlap in both the time and frequency domains. Numerical simulations and experiments were conducted to investigate the FrFT's separation performance for guided waves inspection with array transducers. Results showed that a smaller time delay worsened separation, while using a chirp signal with a broader bandwidth improved separation for signals of fixed duration. Additionally, the effect of signal dispersion on the results was minimal. The findings confirm that the FrFT can effectively separate overlapping signals, enhancing time efficiency in guided wave inspections using array transducers.