一种井下近钻头弱SNR信号的STD-EEMD提取方法  

A STD-EEMD extraction method for the downhole near-bit weak SNR signals

作　　者：毛艳慧 杨龙瀚 霍爱清 李飞[1,2] 高怡 MAO Yanhui;YANG Longhan;HUO Aiqing;LI Fei;GAO Yi(Shaanxi Key Laboratory of Measurement and Control Technology for Oil and Gas Wells,Xi'an 710065,China;College of Electronic Engineering,Xi’an Shiyou University,Xi’an 710065,China)

机构地区：[1]陕西省油气井测控技术重点实验室,西安710065 [2]西安石油大学电子工程学院,西安710065

出　　处：《中国惯性技术学报》2024年第12期1184-1190,共7页Journal of Chinese Inertial Technology

基　　金：国家自然科学基金企业创新发展联合基金重点项目(U20B2029);陕西省自然科学基础研究计划(2023-JC-YB-453)。

摘　　要：实际钻井过程中,底部钻具的强振动以及工具随钻柱的高速旋转等因素,使得工具姿态测量原始信号幅度微弱,信噪比(SNR)极低。针对这一问题,提出了一种基于软阈值集合经验模态分解(STD-EEMD)的近钻头姿态信号提取方法。首先通过最小化偏差寻找EEMD最优分解阶数,然后基于最优分解阶数和相应各阶本征模态函数,利用方差贡献率划分噪声分量与有效分量。接着基于设计的软阈值,对选取的噪声分量进行软阈值滤波处理,进一步提取噪声分量中耦合的有用信号,并基于提取结果与有效分量,实现有效信号的精确重构。加入实际采集的井下数据进行实验验证,结果表明:与DWT、EMD和EEMD方法相比,所提方法的工具面角精度分别提高了65.8%、55.8%、45.3%,井斜角精度分别提高了55.3%、45.0%、36.3%。In the actual drilling process,the strong vibration of the bottom drilling tool and the high-speed rotation of the tool with the drill column make the original signal amplitude of the tool attitude measurement weak and the signal to noise ratio(SNR)extremely low.Aiming at this problem,a soft threshold ensemble empirical mode decomposition(STD-EEMD)extraction method for weak SNR signals of the downhole near-bit is proposed.Firstly,the optimal decomposition order of EEMD is obtained by minimizing the bias,and then the noise component and the effective component are divided by variance contribution rate based on the optimal decomposition order and the corresponding EEMD eigenmode functions.Secondly,based on the designed soft threshold,soft threshold filtering is carried out on the selected noise component,and the coupled useful signals in the noise component are further extracted.The effective signal is finally reconstructed accurately based on the extraction results and effective components.Experimental verification is carried out by adding actual collected downhole data,and the results show that compared with DWT,EMD and EEMD methods,the accuracy of tool face angle of the proposed method is improved by 65.8%,55.8%and 45.3%,and the accuracy of well angle is improved by 55.3%,45.0%and 36.3%,respectively.

关 键 词：导向钻井 姿态测量 集合经验模态分解 软阈值滤波 方差贡献率 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]