摘要
实际管网泄漏检测环境中,管内外非泄漏固定噪声(如施工、管道阀门噪声等)叠加在管道泄漏声信号中,传感观测信号间的互相关函数中有多个相关峰值,若不能正确识别峰值来自泄漏还是非泄漏声源,将导致漏点定位错误。以两传感观测信号的自相关函数构建非泄漏和泄漏信号特征提取函数,通过特征提取函数识别出管道传播媒质对两类信号的衰减特征信息,该衰减特征结合互相关函数的相关峰值时延信息,识别出峰值来自泄漏还是非泄漏声源。实际定位应用表明,在管内外非泄漏固定噪声干扰下,漏点定位误差在1 m以内,突破了传统基于时延估计的泄漏定位技术及仪器需要在夜间,或人为关闭某些管道阀门、支路等安静环境下,通过回避干扰噪声才能进行定位的限制。
Correlation technique is widely used to locate leaks in buried water pipes. If a leak exists, a distinct peak may be found in the cross-correlation function of two acquired acoustic signals. However, in practice, distinct peak in the cross-correlation function may be caused by a non-leak acoustic source occurring in or outside the pipe. And the peak not related to a leak will result in a false location. The autocorrelation functions of the two acquired signals are used to construct a new function. The power attenuation factors of leak and non-leak signals can be extracted from the new function using correlation analysis and approximate entropy. Thus, if the peak in the cross-correlation function of two acquired signals is related to the leak signal or the non-leak signal can be identified. The proposed method has been employed to locate leak in the presence of non-leak noises inside and outside pipes, and the leak location errors are less than 1 m. The proposed method breaks through the limitation of traditional leak location tech- nologies and instruments based on time delay estimation, which can only implement leak location operations at night or in quiet environment where certain pipe valves and slip are intentionally closed.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2012年第2期248-254,共7页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(60804061)
国家重大科技专项(2009ZX07424-004-02)
重庆市科技攻关项目(CSTC
2009AB2041
2010AA3001)
重庆市自然基金(CSTC
2009BB31981)资助项目
关键词
泄漏定位
非泄漏声源
声信道衰减
时延估计
leak location
non-leak acoustic source
acoustic channel attenuation
time delay estimation