Automatic Judgment of Ultrasonic Detection for Solid Axle Inlaid Part

Abstract

Ultrasonic detection is the most effective nondestructive method for axle inlaid part, which has high efficiency and wide coverage. In the process of rail vehicle running, the influence of the external environment and the interaction between wheel and axle cause uneven distribution of stress concentration and with fretting corrosion, which lead to the generation of fatigue crack, otherwise, the waveform of ultrasonic detection is complex, because the energy of non-defect is too high to identify the defect. In view of this phenomenon, the ultrasonic echo data is analyzed by wavelet transform method to extract the defect characteristic, which is used for automatically judgment. The results show that the fatigue defect can be identified effectively by using the defect characteristic that extracted by wavelet transform method.