中文摘要：

针对传统的材料成分分数检测方法需损坏被测材料这一问题,提出一种基于声发射信号衰减特性的材料成分分数检测新方法。选取微晶石墨/聚乙烯醇（PVA）复合材料为对象,从声发射原信号和声发射信号特征小波包两个方面,建立其衰减特性与材料成分分数间的关系,并对该关系进行验证和两方面的验证结果的对比。结果表明：本研究选取信号的特征小波包为125-171.85 kHz频段时,对于原矿微晶石墨/PVA复合材料以此特征小波包计算原矿微晶石墨质量分数误差为2.2%,比用原信号计算误差低1.6%;对于提纯微晶石墨/PVA复合材料以此特征小波包计算提纯微晶石墨质量分数误差为2.4%,比用原信号计算误差低2.0%,检测结果与材料实际成分分数吻合度较高。因此,该方法为材料成分分数无损检测提供了一种新的研究手段。

英文摘要：

Aimed at an issue that materials will be damaged using the traditional methods to detect their composition fractions,a new detection method based on attenuation characteristics of acoustic emission（ AE） signals is provided in this study.Microcrystalline graphite fractions in the microcrystalline graphite / poly vinyl alcohol（ PVA） composite materials were measured.Two relationships were established.One is the relationship between the attenuation coefficients of AE original signals and the microcrystalline graphite fractions.The other is the relationship between the attenuation coefficients of the feature wavelet packet and the microcrystalline graphite fractions.Then the relationships were validated by the examples and their results were compared.The research shows that the frequency range of the feature Wavelet packet is from 125 kHz to 171.85 kHz.For the non-puried microcrystalline graphite / PVA composite,the error of the microcrystalline graphite component fraction calculated by the feature packet energy is 2.2%,which is lower than that of the original signal 1.6%.For the purified microcrystalline graphite / PVA composite,the error calculated by the feature packet energy is 2.4%,which is lower than that of the original signal 2.0%.Themeasured results have the higher degree of agreement with the actual composition fractions.Therefore,the method provides a new research tool for the non-destructive testing of material component scores.