为了通过超高速撞击声发射信号识别蜂窝结构受空间碎片撞击后的损伤状态,提出一种基于小波的损伤特征提取方法.采用超高速撞击声发射技术,以铝合金蜂窝板为研究对象,通过超高速撞击实验获取实验信号.分析超高速撞击声发射信号的时频特征及板波模态等特征,采用Daubechies小波变换将信号中模态分离,根据小波系数计算各尺度小波能量分数及小波能量熵特征,分析各特征参数与损伤间的关系,并通过Kruskal-Wallis检验方法验证各特征值对损伤识别的贡献.结果表明:小波能量分数和小波能量熵具有一定的损伤模式分类能力;250 kHz以上的小波能量分数具有良好的损伤模式分类能力;非超声部分的低频信号对损伤识别存在干扰.%In this work,a hypervelocity impact acoustic emission signal feature extraction method was proposed to detect damages experienced by the honeycomb core sandwich structure impacted by space debris by using hypervelocity impact acoustic emission signals.Varieties of hypervelocity impact acoustic emission signals were obtained through experiments based on the hypervelocity impact acoustic emission on the aluminum honeycomb core sandwich,their time-frequencies and the modes of the waves on the honeycomb plate were analyzed,the modes of the signals were differentiated,and the wavelet energy fraction and entropy were calculated,both by using the Daubechies wavelet decomposition,with the relationship between these parameters and the damage delineated and the contribution of each parameter gauged by the Kruskal-Wallis test.The results show that,to a certain degree,the wavelet energy fraction and the entropy of information are able to identify the damage patterns.Specifically,the energy fraction with a frequency above 250 kHz exhibits a better identifying capability,while signals of a lower frequency out of the ultrasonic range exert disturbance on the damage identification.
展开▼
