玻璃纤维复合材料拉伸损伤的声发射信号模式识别分析  

Acoustic emission signal pattern recognition of tensile damage of glass fiber composite

作　　者：万乐 闫小青[1] 张纯[1] 陈红亮 WAN Le;YAN Xiaoqing;ZHANG Chun;CHEN Hongliang(School of Civil Engineering and Architecture,Nanchang University,Nanchang 330031,China)

机构地区：[1]南昌大学建筑工程学院,江西南昌330031

出　　处：《南昌大学学报（工科版）》2020年第1期23-27,共5页Journal of Nanchang University(Engineering & Technology)

基　　金：国家自然科学基金资助项目(51469016);江西省教育厅科技资助项目(GJJ13053);江西省研究生创新专项资金项目(CX2018057)。

摘　　要：对玻璃纤维复合材料进行常温拉伸破坏实验,利用声发射检测技术实时监测其损伤过程。研究了材料在拉伸破坏过程中的损伤演化规律,发现玻璃纤维复合材料的损伤演化过程可分为损伤发展阶段、损伤加剧阶段和试件失效阶段。同时为了研究材料内部破坏的损伤模式,提取了声发射特征参数上升时间、持续时间、计数、幅值和峰频,结合主成分分析法和K-means聚类的方法对声发射信号进行聚类分析,将聚类的结果与损伤模式相关联,结果表明复合材料损伤模式基体开裂、纤维脱粘和纤维断裂所产生的声发射信号在幅值、计数、持续时间参数上具有不同的响应特征。The tensile failure experiment of glass fiber composite at room temperature was carried out,and the damage process was monitored in real time by using acoustic emission detection technology.The damage evolution law of the material in the tensile failure process was studied.It was found that the damage evolution process of glass fiber composite can be divided into damage development stage,damage aggravation stage and test piece failure stage.At the same time,in order to study the damage mode of material internal failure,the rising time,duration,counting,amplitude and peak frequency of acoustic emission characteristic parameters were extracted.Combined with principal component analysis and K-means clustering method,the acoustic emission signal was cluster analyzed.The clustering results were related to the damage mode.The results showed that the acoustic emission signals produced by matrix cracking,fiber debonding and fiber fracture have different response characteristics in amplitude,counting and duration parameters.

关 键 词：复合材料 声发射 主成分分析 K-MEANS聚类 损伤模式 

分 类 号：TB33[一般工业技术—材料科学与工程]