焊接裂纹金属磁记忆信号的特征提取与应用

Feature extraction of metal magnetic memory signal and its application for weld crack

金属磁记忆检测技术是一种新兴的对铁磁性材料进行早期损伤诊断的无损检测方法。通过对焊接裂纹磁记忆信号的小波能量谱特征进行研究表明,与无裂纹时相比,含有焊接裂纹信息的金属磁记忆信号,其尺度-小波能量谱的分布范围广,能量水平高,且能量达到峰值以后,呈指数规律下降,据此可以判定被测试件中是否含有裂纹缺陷;其空间-小波能量集中程度较无裂纹时高,而能量集中的位置恰好就是焊接裂纹存在的位置。

Metal magnetic memory(MMM) is a new non-destructive testing method which can make a diagnosis for the ferro-magnetic material in early damaging stage.The wavelet energy spectrum feature of MMM signal for weld cracks.Research shows that,for MMM signals with cracks,the distribution scope of scale-wavelet energy spectrum is wide with a high energy level,and the energy spectrum assumes an exponential attenuation after it reaches a peak,which can be used to estimate whether the sample included crack or not.The space-wavelet energy with crack is higher than that withowt crack,Where the space-wavelet energy spectrum indicates a concentration is the very position of the crack.