模糊聚类分析对多层管泄漏声信号的分类识别
详细信息 查看官网全文
- 英文论文题名：Fuzzy clustering analysis for classification and identification of leakage acoustic signals for multilayer pipe
- 论文作者：周博文 ; 胡文祥
- 英文论文作者：ZHOU Bo-wen ; HU Wen-xiang ; Institute of acoustics ; Tongji University
- 年：2016
- 作者机构：同济大学声学研究所;
- 论文关键词：多层管 ; 泄漏检测 ; 模糊聚类 ; 分类识别
- 英文论文关键词：multilayer pipe ; leak detection ; fuzzy clustering ; classification and identification
- 会议召开时间：2016-10-28
- 会议录名称：2016年全国声学学术会议论文集
- 语种：中文
- 分类号：TE973.6
- 学会代码：OGSM
- 会议名称：2016年全国声学学术会议
- 会议地点：中国湖北武汉
- 主办单位：中国声学学会
- 学会名称：中国声学学会
- 页数：4
- 文件大小：317k
- 原文格式：D
- 会议级别：全国
摘要
多层输油管道在连续运行过程中,管道破裂或穿孔等事故时有发生,严重威胁管道安全。因此,当泄漏发生时,能及时检测到泄漏的具体情况是十分必要的。本文针对实验室充液多层压力管道内水听器记录的实验测试数据,利用模糊聚类分析方法,通过模糊C均值动态聚类算法得到最优分类矩阵和聚类中心。根据该方法对不同的泄漏信号进行特征提取,对所提取的特征进行聚类分析,从而达到对多层管的泄漏信号进行分类和识别的目的。该方法能较为快速和准确的判断泄漏信号的压力范围和泄漏口尺寸大小范围以及定位其所属管层等关键信息。实验信号的处理结果,证明了此方法的有效性和可行性。
It is possible that multilayer pipe may burst, leak, or other accidents may happen in the process of continuous operation. It is a seriously threat to the safe running of pipe system. Therefore, it is necessary to detect the leakage in time when it occurs. In this paper, the fuzzy clustering analysis method is used to get the optimal classification matrix and clustering center by the fuzzy C mean dynamic clustering algorithm. According to this method, using experimental testing leaky noise signals acquiring from the hydrophone inside the multilayer pressure pipe filled with water in laboratory, the different leakage signals are extracted to achieve the classification and identification. The results show that it is rapid and accurate to judge the key information such as pressure range of the pipe, the leaky hole size, and the location which the layer the hole belong to. The effectiveness and feasibility of the method are proved by the experimental signal processing.
It is possible that multilayer pipe may burst, leak, or other accidents may happen in the process of continuous operation. It is a seriously threat to the safe running of pipe system. Therefore, it is necessary to detect the leakage in time when it occurs. In this paper, the fuzzy clustering analysis method is used to get the optimal classification matrix and clustering center by the fuzzy C mean dynamic clustering algorithm. According to this method, using experimental testing leaky noise signals acquiring from the hydrophone inside the multilayer pressure pipe filled with water in laboratory, the different leakage signals are extracted to achieve the classification and identification. The results show that it is rapid and accurate to judge the key information such as pressure range of the pipe, the leaky hole size, and the location which the layer the hole belong to. The effectiveness and feasibility of the method are proved by the experimental signal processing.
引文
[1]高新波.模糊聚类分析及其应用.西安电子科技大学出版社,2004,37-39.
[2]L.A.Zadeh,Fuzzy sets,Information and control.1965,8(3):338-353.
[3]沈清,汤霖,模式识别导论,北京:国防科技大学出版社,1991,126~127.
[4]李弼程,邵美珍,黄洁.模式识别原理与应用.西安电子科技大学出版社,2008,144-163.
[5]Yunjie Zhang,Weina Wang,Xiaona Zhang,Yi Li.A cluster validity index for fuzzy clustering.Information Sciences,Volume178,Issue4,15February 2008,1205-1218.
[6]崔谦.油气管道泄漏检测方法的研究及应用[D].2005.87-88.
[2]L.A.Zadeh,Fuzzy sets,Information and control.1965,8(3):338-353.
[3]沈清,汤霖,模式识别导论,北京:国防科技大学出版社,1991,126~127.
[4]李弼程,邵美珍,黄洁.模式识别原理与应用.西安电子科技大学出版社,2008,144-163.
[5]Yunjie Zhang,Weina Wang,Xiaona Zhang,Yi Li.A cluster validity index for fuzzy clustering.Information Sciences,Volume178,Issue4,15February 2008,1205-1218.
[6]崔谦.油气管道泄漏检测方法的研究及应用[D].2005.87-88.