Bootstrap与模糊聚类在管道泄漏检测中研究  

Research on Bootstrap and Fuzzy Cluster Method in Pipeline Leak Detection

作　　者：李炜[1] 苗丽[1] 张耿[1] 

机构地区：[1]兰州理工大学电气工程与信息工程学院,甘肃兰州730050

出　　处：《微计算机信息》2007年第31期292-294,共3页Control & Automation

基　　金：教育部春晖计划(Z2005-1-62001)

摘　　要：管道泄漏信息缺失对可靠提取小泄漏与工况变化所产生的负压波模式的特征一直是一瓶颈,也是降低系统误报警率的关键因素之一。本文利用小样本统计的Bootstrap方法,对观测样本进行模拟采样并进行统计分析,在此基础上,对管道不同工况提取特征,并用模糊聚类分析的方法分类管道小泄漏与工况变化的运行模式。仿真实验表明该方法应用在管道泄漏诊断中是简单实用和可行的。Pipeline leakage detection with Incomplete Information is always a bottleneck to reliable extract the feature of negative pressure wave mode producing by small leak and the different work conditions, which is one of the key factors that reducing the system false alarm rate. This paper utilized Bootstrap method based on the small sample statistics, sampling simulation and the statistical analysis of observation sample which is also made, features of pipeline in different work conditions are extracted, utilizing the fuzzy cluster analysis method to classify the operation mode of pipeline small leak and different work conditions. The simulation results are verit^ed that the method is simple, feasible and effective.

关 键 词：管道 泄漏检测 统计模拟 特征提取 模糊聚类 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]