机构地区: 华南理工大学机械与汽车工程学院
出 处: 《现代制造工程》 2006年第4期90-92,共3页
摘 要: 超声无损检测一直是无损检测领域研究的热点,由于材料内部组织结构的复杂性,超声回波信号中往往存在着大量的噪声干扰。引入一种新的处理非线性、非平稳信号的经验模式分解法(EMD)来处理超声检测信号。该方法把检测信号分解若干个内在模式分量(IMF),突出了信号的局部特征。结果表明,EMD方法能有效抑制干扰,对于360mm深的缺陷试样,缺陷定位精度达1mm。 Ultrasonic nondestructive testing is always a focus in the nondestructive testing domain. A lot of noise exists in the ultrasonic echo wave because of the complicated internal structure of material. A method suitable for non-linear and non-stationary signal is proposed to process the ultrasonic test signal. The diagnosis signal is decomposed into several Intrinsic Mode Functions (IMF) by Empirical Mode Decomposition (EMD). The results show that the approach based on EMD can restrain noise effectively and advance sample's flaw location accuracy to 1mm.