【作者】 周海旭； 叶昊； 王桂增；

【Author】 Zhou Haixu Ye Hao Wang Guizeng (Dept.of Automation Tsinghua Univ Beijing 100084 China)

【机构】 清华大学自动化系；

【摘要】 本文研究了基于声波传感器的输气管道泄漏检测问题,分别提出了基于模糊C均值聚类的自适应泄漏可能性评价方法,及基于D-S证据理论的多传感器融合泄漏检测方法。基于现场数据的离线测试表明,该方法可以成功检测出泄漏,并可在一定程度上减小部分传感器误报或漏报所导致的影响,提高了泄漏检测的可靠性。更多还原

【Abstract】 The problem of leak detection of gas pipeline based on acoustic sensors is discussed in this paper.The paper proposes an adaptive leakage possibility evaluation method based on fuzzy C-means clustering,and a multi-sensor fusion leakage detection method based on D-S evidence theory respectively.Offline tests based on field data show that,the approach can not only detect leakage successfully,but also decrease the influence caused by false alarm or miss detection of parts of sensors,and thus can improve the reliability of the detection.更多还原