基于模糊分类的流体管道泄漏故障智能检测方法研究  

Research on leak fault intelligent detection method for fluid pipeline based on fuzzy classification

作　　者：刘金海[1] 冯健[1] 

机构地区：[1]东北大学信息科学与工程学院,沈阳110004

出　　处：《仪器仪表学报》2011年第1期26-32,共7页Chinese Journal of Scientific Instrument

基　　金：中央高校基本科研业务费(N100304007);国家高技术研究发展计划(863计划)(2009AA04Z127);国家自然科学基金(50977008);新世纪优秀人才支持计划(NCET-08-0101)资助

摘　　要：本文针对基于负压波法管道泄漏实时检测系统误报高和灵敏度低的问题提出一种流体管道泄漏故障智能检测方法,该方法首先给出管道运行参数的确定模型,然后结合模糊算子给出流体管道状态模糊模型,进而利用该模型实现管道故障分类。以这种智能检测方法为核心设计流体管道故泄漏故障智能诊断系统(leak intelligent diagnosis system for fluid pipeline,LIDSFP),通过对某成品油管道实例仿真和在流体管道测试系统上的试验研究,给出了LIDSFP性能指标,进一步分析表明该系统可以有效完成流体管道的泄漏故障诊断。A leak fault intelligent detection method for fluid pipeline based on fuzzy classifier is proposed,which can decrease false alarms and improve leak detection sensitivity.To complete real-time and exact fault diagnosis of fluid pipeline,a fuzzy classifier for operation states is designed according to the framework of fuzzy expert system.A leak fault intelligent diagnosis system for fluid pipeline（LIDSFP） was designed with this intelligent method;simulation was carried out to detect faults from the historic operation data of certain petrolatum product fluid pipeline in China.Test study was also carried out on the proposed system and the performance specification of the system is given.Simulation and test results show that the proposed diagnosis system has good performance.

关 键 词：流体管道 泄漏检测 模糊分类 

分 类 号：TE88[石油与天然气工程—油气储运工程] TP277[自动化与计算机技术—检测技术与自动化装置]