基于ICEEMDAN和加权融合算法的多传感器输油管道泄漏定位方法研究  

Research on Leak Location Method of Multi-sensor Oil Pipeline Based on ICEEMDAN and Weighted Fusion Algorithm

作　　者：陈伟[1] CHEN Wei(Huabei Oilfield Company,CNPC)

机构地区：[1]中国石油天然气集团公司华北油田分公司

出　　处：《油气田地面工程》2025年第2期49-54,64,共7页Oil-Gas Field Surface Engineering

基　　金：中国石油股份有限公司重大专项“华北油田持续有效稳产勘探开发关键技术研究与应用”(2017E-15)。

摘　　要：针对负压波泄漏技术中数据利用率低、信号背景噪声大、数据融合度差等问题,提出了一种基于ICEEMDAN和加权融合算法的多传感器管道泄漏定位方法。首先,对三传感器系统的泄漏定位原理进行梳理,分析负压波波速差异性带来的误差;其次,通过添加成对高斯白噪声对CEEMDAN算法进行改进,以期实现负压波信号的降噪;最后,以异常值检测为基础,通过分组、分批加权实现多时间窗口数据的融合。结果表明,三传感器系统有利于辨识负压波是否由泄漏所引起,与负压波波速差异性相比,定位结果的最大相对误差仅为9.09%;ICEEMDAN算法在收敛速度和降噪效果上均具有优越性,不同传感器位置上的重构信号信噪比在12.9987~13.6745之间,均方根误差在5.2106~5.8433之间,优于CEEMDAN降噪算法;加权融合后时差数据ΔtAB和ΔtAC的标准差分别为0.008和0.009,较原始数据降了两个数量级,泄漏定位的绝对误差为35.69 m,相对误差为1.74%,与算术平均法和单一数据法相比,定位精度大幅提升。该研究结果可为管道泄漏检测及维抢修工作的开展提供实际参考。Aiming at the problems of low data utilization,large signal background noise,and poor data integration in the existing negative pressure wave leakage technology,a multi-sensor pipeline leak location method based on ICEEMDAN and weighted fusion algorithm is proposed.Firstly,the principle of leakage location of the three-sensor system is sorted out,and the error caused by the difference in negative pressure wave velocity is analyzed.Secondly,the CEEMDAN algorithm is improved by adding pairs of Gaussian white noise to realize the noise reduction of negative pressure wave signals.Finally,on the basis of outlier detection,multitime window data fusion is realized by grouping and batch weighting.The results show that the three-sensor system is beneficial in identifying whether the negative pressure wave is caused by leakage,and the maximum relative error of the positioning results is only 9.09%compared with the difference of the negative pressure wave velocity.The ICEEMDAN algorithm has advantages in both convergence speed and noise reduction effect.The reconstructed signal SNR of different sensor positions ranges from 12.9987 to 13.6745,and the root-meansquare error ranges from 5.2106 to 5.8433,which is superior to CEEMDAN algorithm.After weighted fusion,the standard deviations ofΔtAB andΔtAC are 0.008 and 0.009,respectively,which are two orders of magnitude lower than the original data.The absolute error of leakage location is 35.69 m,and the relative error is 1.74%.Compared with the arithmetic average method and single data method,the positioning accuracy has been greatly improved.The research results can provide practical reference for pipeline leakage detection and repair work.

关 键 词：ICEEMDAN 负压波 加权融合 定位 管道泄漏 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]