摘要
针对粘接结构的声-超声检测信号,提出用短时傅里叶变换的方法对缺陷进行识别。该方法依据短时傅里叶变换特性,合理选择窗函数及其宽度,并根据时频特征值对识别结果进行体现。结果表明:试件中缺陷的存在与否对时频特性影响较大,而相同性质缺陷的检测信号时频特性相近;脱粘缺陷信号具有相对较好的频率分辨率,而孔洞缺陷信号具有较好的时间分辨率。该方法对声-超声检测信号进行缺陷识别是有效、可行的。
Aimming at acousto-ultrasonic signals defect identification of adhesive-bonded structure,the short-time fourier transform(STFT) method was proposed in this paper. This method,which is according to characteristics of the STFT, selected window function and its width reasonably, and showed the identification results by time-frequency characteristics. As a result,there are large differences on the time-frequency characteristics of the test signals if the structure has flaws,and the similar if the flaws are the same.Bonding defects have good frequency resolution,while hole defects have good temporal resolution.This method is effective and feasible for defect identification of acousto-ultrasonic test signals.
出处
《中国测试》
CAS
北大核心
2015年第4期29-31 35,35,共4页
China Measurement & Test
