改进LMD及高阶模糊度函数的管道泄漏定位

Pipeline leakage localization based on ELMD and high-order ambiguity function

针对天然气管道泄漏定位的问题,提出一种基于改进局域均值分解(LMD)及高阶模糊度函数的时延估计方法。该方法首先采用改进的LMD对声发射信号进行分解,获得多个PF分量,进而提出根据K-L散度的PF分量自动选择算法,获取含有主要泄漏信息的PF分量,在此基础上,研究了基于高阶模糊度函数计算声发射信号的时频参数,并通过时频分析获取特征频率的到达时间差,最后结合泄漏产生的广义声发射信号的传播速度完成对天然气管道泄漏的定位。实验结果表明,提出的方法能够进行定位且精度较直接相关法明显提高。

Aiming at the localization problem in natural gas pipeline leakage detection,a time delay estimation method based on ensemble local mean decomposition（ ELMD） and high-order ambiguity function is proposed. Firstly,the ELMD method was adopted to decompose the acoustic emission signals and obtain the production function（ PF） components. A PF component automatic selection method based on K-L divergence was used to process these PF components and obtain the principal PF components containing most of the leakage information. On this basis,the high-order ambiguity function was applied to calculate the time-frequency parameters of the principal PF components,and obtain the arrival time difference of the characteristic frequencies through time-frequency analysis. Finally,combining the time difference with the signal propagation speed of the generalized acoustic emission signals produced by the leakage,the natural gas pipeline leakage localization was accomplished. Experiment results show that the proposed method can locate the leakage,and the localization accuracy is apparently higher than that of the direct correlation method.