基于声-压信号融合的城市供水管道泄漏监测
DOI:
CSTR:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

TU991. 36 TN911. 7 TH86

基金项目:

国家自然科学基金(52378525)项目资助


Leakage monitoring of urban water supply pipelines based on fusion of acoustic-pressure signals
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    高效监测城市供水管道漏损对于节约水资源和保障居民用水安全具有重要意义。 现有泄漏检测方法主要依赖于单一 信号,单一信号因其自身局限性或对小泄漏引起的微小波动不敏感或易受管道正常运行行为干扰,无法通过识别方法层面解 决。 研究基于声学信号和压力信号各自优势,建立基于声-压信号融合的泄漏监测策略,有效解决了声信号误报率高与压力信 号漏报率高的问题,提高了检测系统精度。 在某长距离在役供水管道进行泄漏模拟原位测试验证泄漏监测策略的识别效果并 讨论两类流体内参数信号的可识别距离。 经验证,所提策略使误报率降低 6. 02% ,漏报率降低 4. 57% 。

    Abstract:

    Efficient monitoring of leakage in urban water supply pipelines is of paramount importance for conserving water resources and ensuring the safety of residential water use. Existing leakage detection methods mainly rely on a single type of signal. Due to the inherent limitations of single signals, they are either insensitive to the minor fluctuations caused by small leaks or susceptible to interference from normal pipeline operations, which cannot be resolved through identification methods alone. This study exploits the respective advantages of acoustic and pressure signals and establishes a leakage monitoring strategy based on fusion of acoustic-pressure signals, effectively addressing the issue of high false positives in acoustic signals and high false negatives in pressure signals. The proposed strategy was validated through in-situ leak simulation tests on a long-distance operational water supply pipeline, evaluating the detection effectiveness and discussing the identifiable distance for both types of fluid parameter signals. The results demonstrate that the proposed strategy reduces the false alarm rate by 6. 02% and the miss detection rate by 4. 57% . Keywords:water supply pipeline; internal pipeline parameters monitoring; ada

    参考文献
    相似文献
    引证文献
引用本文

张 祎,李素贞.基于声-压信号融合的城市供水管道泄漏监测[J].仪器仪表学报,2024,45(8):69-76

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2024-12-18
  • 出版日期:
文章二维码