论文部分内容阅读
针对固体火箭发动机壳体/绝热层界面缺陷进行超声定量检测问题,本文提出采用经验模态分解(EMD)对信号进行降噪处理。通过对信号进行EMD处理,求得各阶本征模态函数(IMF)。通过分析各阶IMF分量的时域信号、频域信号与原始信号的相关性,提出将IMF3分量作为缺陷定量的特征信号,采用峰值引力波因子(SWF)作为特征量来评价特征信号,通过比较SWF数值大小实现了粘接结构脱粘缺陷的定量检测。
In order to solve the problem of ultrasonic quantitative detection of the solid rocket motor housing / insulation interface defects, this paper proposes using empirical mode decomposition (EMD) to denoise the signal. Through the EMD signal processing, to obtain each order intrinsic mode function (IMF). By analyzing the correlation between the time-domain signals and the frequency-domain signals of the IMFs in different orders, we propose to use the IMF3 component as the quantitative feature of the defect and the peak gravity wave factor (SWF) as the feature to evaluate the characteristic signals. By comparing SWF numerical size to achieve the adhesive structure debonding quantitative detection.