基于K均值聚类的FRP复合材料容器损伤声发射信号模式识别  

Pattern Recognition of Acoustic Emission Signals of FRP Composite Vessel Damage Based on K- means Clustering

作　　者：李伟[1] 付玉[1] 蒋鹏[1] 李英年[1] 

机构地区：[1]东北石油大学机械科学与工程学院,黑龙江大庆163318

出　　处：《压力容器》2014年第8期14-19,共6页Pressure Vessel Technology

基　　金：国家科技支撑计划(2012BAH28F03);博士后研究人员落户黑龙江科研启动资助项目(LBHQ10177)

摘　　要：采用声发射技术对FRP复合材料容器爆破过程进行监测,应用K均值聚类方法对预处理后的声发射信号进行分类,结合声发射信号的参数分析实现损伤机制的模式识别,并对不同损伤机制的声发射特性进行对比研究。研究结果为FRP复合材料损伤评价与识别提供了依据。The acoustic emission(AE)technology was used to monitor the blasting process of fiber rein-forced polymer(FRP)composite vessels.The K means clustering algorithm was used to classify the pre-treatment of the AE signals.With parameter analysis of AE signal,pattern recognition of fatigue damage was accomplished,and we also made some comparison study with emission property of diffrent fatigue damages.The results will provide theoretical basis for the identification and evaluation of FRP.

关 键 词：FRP复合材料 声发射 K均值聚类 模式识别 

分 类 号：TB302[一般工业技术—材料科学与工程] TB33