改进VMD与广义互相关在供水管道泄漏定位中的应用  

Improved VMD and generalized cross-correlation for leakage localization in water supply pipelines

作　　者：胡佳佳 魏媛媛 付世沫 王耀力[1] 常青[1] HU Jiajia;WEI Yuanyuan;FU Shimo;WANG Yaoli;CHANG Qing(College of Electronic Information and Optical Engineering,Taiyuan University of Technology,Jinzhong 030600,China;Taiyuan Water Supply Design and Research Institute Co.,Ltd.,Taiyuan 030001,China)

机构地区：[1]太原理工大学电子信息与光学工程学院,山西晋中030600 [2]太原供水设计研究院有限公司,山西太原030001

出　　处：《现代电子技术》2024年第19期75-82,共8页Modern Electronics Technique

基　　金：山西省重点研发项目(201903D321003);太原供水设计研究院有限公司项目(RH2000005391);山西省自然科学基金项目(201801D121141)。

摘　　要：针对管道泄漏定位时,采集到的泄漏信号会受到噪声干扰而影响定位精度的问题,提出了一种参数自寻优变分模态分解结合广义互相关的定位方法。首先,在现有的频域加权函数的基础上,提出了一种复合加权函数,对不同的频域加权函数在供水管道泄漏声信号下的性能进行研究;其次,利用高斯变异和量子行为对粒子群进行优化,建立复合指标排列熵/互信息熵作为适应度函数,改善粒子在求解过程中的多样性;然后,通过改进的粒子群算法优化变分模态分解(VMD)最优参数组合,实现对信号的降噪处理。实验结果表明,文中方法在供水管道泄漏情况下具有更高的定位精度,平均相对定位误差仅为3.7%,且最大误差也只有8.0%,能够有效地滤除原始信号中的噪声干扰。When locating a pipeline leakage,the collected leakage signals are susceptible to noise interference,which will affect the positioning accuracy in turn,so a positioning method of parameter self-optimizing variational modal decomposition(VMD)combined with generalized cross-correlation(GCC)is proposed.A composite weighting function is proposed on the basis of the existing frequency-domain weighting functions.The performance of different frequency-domain weighting functions during the occurrence of the leakage acoustic signals is studied.The Gaussian mutation and quantum behavior are used to optimize the particle swarm,and a composite index of permutation entropy/mutual information entropy is established and taken as the fitness function to improve the diversity of particles during the process of solving.The optimal parameter combination of VMD is optimized by the improved particle swarm optimization(PSO)algorithm,so that the noise reduction of the signals is achieved.The experimental results demonstrate that the proposed method exhibits higher localization accuracy in the case of water supply pipeline leakage,with an average relative positioning error of merely 3.7%,and a maximum error of only 8.0%,so the method can eliminate noise interference from the original signals effectively.

关 键 词：供水管道 泄漏定位 粒子群优化算法 变分模态分解 时延估计 广义互相关 

分 类 号：TN911.4-34[电子电信—通信与信息系统]