| Ultrasonic nondestructive testing plays an important role in modern industry, it has beenwidely used in stages of the product design, manufacture and using. It can effectively detectand control the quality of the products, so it greatly improves the economic efficiency. Sincethe linear frequency modulation signal has the characteristics of large time-bandwidth product,it can achieve higher resolution and a wider detection range as the transmit signal, which makesit get more and more attention in the field of ultrasonic testing. The traditional detection signal,the pulse signal, has a very short duration, where direct wave interference can be easilyeliminated by time-division way; While research in this paper is based on the linear frequencymodulation signal that has the long duration, where time-division way is no longer applicable,thus effectively eliminating the interference of direct wave is significant to the detectionaccuracy. Based on this issue, this thesis does the research for direct wave separation technology,mainly including the following two aspects:1. Direct wave separation method based on fractional Fourier transform is proposed:Because the fractional Fourier transform of the chirp signal has good energy aggregation, andthe direct wave signal is similar to the transmit signal, it can detect and separate the direct wavein fractional Fourier transform domain.2. For the side lobes shadowing problem of the direct wave in the fractional Fouriertransform domain, the thesis proposes two solutions: based on window functions and based onLMS adaptive filtering. In conventional Fourier transform, the window function method caneffectively suppress the side lobes. The thesis proves the common window functions can alsosuppress the side lobes in fractional Fourier transform domain through numerical simulation.While LMS adaptive filtering algorithm is widely used in time-domain noise reduction, thisthesis takes the direct wave in fractional Fourier transform domain as the interference. With thecombination of LMS algorithm, the simulation result shows that the algorithm can effectivelyremove the direct wave.Finally, with the ultrasonic testing platform in our laboratory, the experiment based on thewindow function algorithm of direct wave separation was conducted, proving the feasibilityand practicality of the algorithm. |