基于K均值聚类的海洋平台T型管节点损伤声发射信号的模式识别

基金项目:

国家科技支撑计划资助项目(2012BAH28F03)

详细信息

Pattern Recognition of AE Signals of Offshore Platform T-tube Damage Based on K-means Clustering

摘要: 对海洋石油平台T型管节点损伤进行了声发射监测,并对损伤检测过程的声发射信号进行K均值聚类分析,以声发射参量的相关性为依据选取损伤特征参数,实现了对T型管节点不同损伤阶段信号的有效识别,为T型管节点的损伤监测和不同损伤阶段的模式识别提供了依据.
- T型管 /
- 声发射 /
- K均值聚类 /
- 模式识别
Abstract: An experiment was performed on the offshore platform T-tube node for the purpose of understanding its damage mechanism.Characteristics of acoustic emission signal of the offshore platform T-tube node at different damage stages were obtained by using the K-means clustering analysis.The damage characteristic parameters of the acoustic emission signal were first selected based on the correlation analysis.The proposed method could thus effectively realize the recognition of T-type pipe node at different damage stages.The study should provide reference for damage monitoring of T-type pipe node and the identification of its different damage stages.
- T-tube /
- Acoustic emission /
- K-means clustering /
- Pattern recognition

[1]	戴光.声发射检测技术在中国——庆祝中国机械工程学会无损检测分会成立三十周年[J].无损检测,2008,30(7):389-394.
[2]	周世兵,徐振源,唐旭清.新的K-均值算法最佳聚类数确定方法[J].计算机工程与应用,2010,46(16):27-31.
[3]	李晶皎,王爱侠,王娇,等.模式识别[M].北京:电子工业出版社,2010:431-432.
[4]	POLLOCK A A.Acoustic emission inspection.ASM handbook,Vol 17.Nondestructive evaluation and quality control [M].Materials Park,USA:ASM International,1989:278-294.