基于频率域经验模式分解阈值滤波的核磁共振测井信号去噪
|
摘要
核磁共振测井信号易受到地层环境和测井设备的干扰,使得测井信号信噪比低下,严重影响后续T_2谱的反演和储层参数的计算.将经验模式分解与阈值滤波技术结合,首先通过Fourier变换将含噪的测井信号变换到频率域,再对实部和虚部分别进行经验模式分解后的阈值滤波,最后将去噪后的实部和虚部分量组合成新的复数序列,通过Fourier反变换恢复时域的测井信号,达到压制噪声的目的.应用该方法对仿真和核磁共振测井信号进行了分析,并与小波软阈值去噪结果进行了对比.结果表明:本文方法在消除噪声的同时,能很好地保留核磁共振测井信号的波形和特征,相比小波软阈值滤波方法,去噪后噪声抑制比由18.17提高到了24.33,幅值衰减比由21.54下降到了17.86,很好地保留了峰形特征,突出了反演的T_2曲线中被淹没的有用信息,为孔隙度、可动流体等储层参数的精确计算提供了保障,为核磁共振测井信号的去噪开拓了新思路.
Nuclear Magnetic Resonance(NMR) logging signals are easily disturbed by noise come from formation environment and logging equipment. Noise make the signal-to-noise ratio of logging signals be low, which seriously influences the inversion of subsequent T_2 spectra and the calculation of reservoir parameters. The empirical mode decomposition is combined with threshold filtering technique. Firstly, the noisy logging signals are transformed into frequency domain by Fourier transform. The threshold filtering is performed after the empirical mode decomposition is applied to the real part and the imaginary part. Then, the de-noised real part and imaginary part are combined into a new complex sequence, and the inverse Fourier transform is used to recover the logging signals in the time domain to achieve the purpose of suppressing noise. The method is used to analyze the simulated and measured NMR logging signals, and is compared with the wavelet soft threshold de-noising method. The results show that the method can keep the characteristics of the echo string of NMR logging signal while eliminating the noise. Compared with the wavelet soft threshold filtering method, the noise suppression ratio is increased from 18.17 to 24.33, and the amplitude attenuation ratio is decreased from 21.54 to 17.86. It is good to keep the peak features and highlights the useful information which is submerged in the inversion of T_2 curve. It provides a guarantee for the accurate calculation of reservoir parameters such as porosity and movable fluid, and opens up a new way for the de-noising of NMR logging signal.
Nuclear Magnetic Resonance(NMR) logging signals are easily disturbed by noise come from formation environment and logging equipment. Noise make the signal-to-noise ratio of logging signals be low, which seriously influences the inversion of subsequent T_2 spectra and the calculation of reservoir parameters. The empirical mode decomposition is combined with threshold filtering technique. Firstly, the noisy logging signals are transformed into frequency domain by Fourier transform. The threshold filtering is performed after the empirical mode decomposition is applied to the real part and the imaginary part. Then, the de-noised real part and imaginary part are combined into a new complex sequence, and the inverse Fourier transform is used to recover the logging signals in the time domain to achieve the purpose of suppressing noise. The method is used to analyze the simulated and measured NMR logging signals, and is compared with the wavelet soft threshold de-noising method. The results show that the method can keep the characteristics of the echo string of NMR logging signal while eliminating the noise. Compared with the wavelet soft threshold filtering method, the noise suppression ratio is increased from 18.17 to 24.33, and the amplitude attenuation ratio is decreased from 21.54 to 17.86. It is good to keep the peak features and highlights the useful information which is submerged in the inversion of T_2 curve. It provides a guarantee for the accurate calculation of reservoir parameters such as porosity and movable fluid, and opens up a new way for the de-noising of NMR logging signal.
引文
Cai J H, Chen QY. 2016. De-noising for NMR oil well logging signals based on empirical mode decomposition and independent component analysis [J]. Arabian Journal of Geosciences, 9: 55- 66.
Cai Jianhua, Xiao Xiao. 2015. Method of Processing Magnetotelluric Signal Based on the Adaptive Thre shold Wavelet [J]. Progress in Geophysics (in Chinese), 30(6): 2433-2439, doi: 10.6038/pg20150601.
Donoho D L. 1995. De-noising by soft-thresholding [J]. IEEE Transaction on Information Theory, 41(03): 613- 627.
Duffy D G. 2004. The application of Hilbert-Huang transforms to meteorological data sets [J]. J. Atmos. Oceanic Technol., 21, 599- 611.
Edwards C. 1996. Improved NMR well logs from Time-dependent echo filtering [C]. in 37th annual logging symposium transactions: Society of Professional Well Log Analysts, 12.
Flandrin P, Rilling G, Goncalves P. 2003. Empirical mode decomposition as a filter bank [J]. IEEE Sig. Process. Letters, 11(2): 112-114.
Freedman R, Morriss CE. 1995. Processing of data from an NMR logging tool [J]. Soc. Pet. Eng. J., 10: 301-316.
Gallegos D P, Smith D M. 1998. A NMR technique for the analysis of pore structure: determination of continuous pore size distributions [J]. J. Colloid Interface Sci., 122(1): 143-153.
GUO Jiliang, LI Hongbing, LI Ming, et al. 2016. Four parameters porosity inversion method representing the effect of pore morphology [J]. Geophysical Prospecting for Petroleum (in Chinese), 55(4): 576-586.
Huang N E, Shen Z, Long S R, et al. 1998. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis [J]. Proc. R. Soc. Lond Ser. A, 454, 903-95.
Huang N E, Wu M C, Long S R, et al. 2003. A confidence limit for the empirical mode decomposition and Hilbert spectral analysis. Proc. R. Soc. Lond Ser A. 459(2037): 2317- 45.
LUO Xin-gang, WANG Wan-yin. 2017. Study of contrast in processing and transformation of field data based on cosine transform and Fourier transform [J]. Progress in Geophysics (in Chinese), 32(3): 1197-1207, doi: 10.6038/pg20170333.
Munn K, Smith D M. 1987. A NMR technique for the analysis of pore structure: numerical inversion of relaxation measurements [J]. J. Colloid Interface Sci., 19(1): 117-126.
QIN Xuan, CAI Jianchao, LIU Shaoyong, et al. 2017. Microseismic data denoising method based on EMD mutual information entropy and synchrosqueezing transform [J]. Geophysical Prospecting for Petroleum (in Chinese), 56(5): 658- 666.
Ren Z, Chen C, Pan K, et al. 2017. Gravity Anomalies of Arbitrary 3D Polyhedral Bodies with Horizontal and Vertical Mass Contrasts [J]. Surveys in Geophysics, 38(2): 479-502.
Suto N, Harada M, Izutsu, J et al. 2006. Time variation of the electromagnetic transfer function of the earth estimated by using wavelet transform[J]. Proc. Jpn. Acad., Ser. B, 82(5): 175-180.
Trad D O, Travassos J M. 2000. Wavelet filtering of magnetotelluric data [J]. Geophysics, 65(2): 482- 491.
WANG Yanwei, XIA Kewen, NIU Wenjia, et al. 2017. A denoising method by low-rank matrix recovery and its application in oil well logging [J]. Geophysical Prospecting for Petroleum (in Chinese), 56(5): 644- 650.
Weng Ai-hua, Li Zhou-bo, Mo Xiu-wen. 2003. Processing of low signal/ noise ratio NMR logging data [J]. Journal of Ji Lin university (earth science edition) (in Chinese), 33(2): 232-235.
Wu Lei, Kong Li, Cheng Jing-jing. 2011. Signal De-noising Algorithm Design in NMR Logging Based on Wavelet Transform [J]. Instrument Technique and Sensor (in Chinese), 10: 71-73, 83.
Xiao Li-zhi, Xie Ran-hong. 2003. Applications of NMR to oil well logging and formation evaluation [J]. Engineering Science (in Chinese), 5(9): 87-94.
蔡剑华, 肖晓. 2015. 基于小波自适应阈值去噪的MT信号处理方法[J]. 地球物理学进展, 30(6): 2433-2439, doi: 10.6038/pg20150601.
郭继亮, 李宏兵, 李明, 等. 2016. 一种体现孔隙形态影响的四参数孔隙度反演方法[J]. 石油物探, 55(4): 576-586.
罗新刚, 王万银. 2017. 余弦变换和Fourier变换在位场数据处理与转换中的对比研究[J]. 地球物理学进展, 32(3): 1197-1207, doi: 10.6038/pg20170333.
秦晅, 蔡建超, 刘少勇, 等. 2017. 基于经验模态分解互信息熵与同步压缩变换的微地震信号去噪方法研究[J]. 石油物探, 56(5): 658- 666.
王艳伟, 夏克文, 牛文佳, 等. 2017. 基于低秩矩阵恢复的去噪方法在石油测井中的应用[J]. 石油物探, 56(5): 644- 650.
翁爱华, 李舟波, 莫修文. 2003. 低信噪比核磁共振测井资料的处理技术[J]. 吉林大学学报(地球科学版), 33(2): 232-235.
吴磊, 孔力, 程晶晶. 2011. 基于小波变换的核磁共振测井信号去噪算法设计[J]. 仪表技术与传感器, 10: 71-73, 83.
肖立志, 谢然红. 2003. 核磁共振在石油测井与地层油气评价中的应用[J]. 中国工程科学, 5(9): 87-94.
Cai Jianhua, Xiao Xiao. 2015. Method of Processing Magnetotelluric Signal Based on the Adaptive Thre shold Wavelet [J]. Progress in Geophysics (in Chinese), 30(6): 2433-2439, doi: 10.6038/pg20150601.
Donoho D L. 1995. De-noising by soft-thresholding [J]. IEEE Transaction on Information Theory, 41(03): 613- 627.
Duffy D G. 2004. The application of Hilbert-Huang transforms to meteorological data sets [J]. J. Atmos. Oceanic Technol., 21, 599- 611.
Edwards C. 1996. Improved NMR well logs from Time-dependent echo filtering [C]. in 37th annual logging symposium transactions: Society of Professional Well Log Analysts, 12.
Flandrin P, Rilling G, Goncalves P. 2003. Empirical mode decomposition as a filter bank [J]. IEEE Sig. Process. Letters, 11(2): 112-114.
Freedman R, Morriss CE. 1995. Processing of data from an NMR logging tool [J]. Soc. Pet. Eng. J., 10: 301-316.
Gallegos D P, Smith D M. 1998. A NMR technique for the analysis of pore structure: determination of continuous pore size distributions [J]. J. Colloid Interface Sci., 122(1): 143-153.
GUO Jiliang, LI Hongbing, LI Ming, et al. 2016. Four parameters porosity inversion method representing the effect of pore morphology [J]. Geophysical Prospecting for Petroleum (in Chinese), 55(4): 576-586.
Huang N E, Shen Z, Long S R, et al. 1998. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis [J]. Proc. R. Soc. Lond Ser. A, 454, 903-95.
Huang N E, Wu M C, Long S R, et al. 2003. A confidence limit for the empirical mode decomposition and Hilbert spectral analysis. Proc. R. Soc. Lond Ser A. 459(2037): 2317- 45.
LUO Xin-gang, WANG Wan-yin. 2017. Study of contrast in processing and transformation of field data based on cosine transform and Fourier transform [J]. Progress in Geophysics (in Chinese), 32(3): 1197-1207, doi: 10.6038/pg20170333.
Munn K, Smith D M. 1987. A NMR technique for the analysis of pore structure: numerical inversion of relaxation measurements [J]. J. Colloid Interface Sci., 19(1): 117-126.
QIN Xuan, CAI Jianchao, LIU Shaoyong, et al. 2017. Microseismic data denoising method based on EMD mutual information entropy and synchrosqueezing transform [J]. Geophysical Prospecting for Petroleum (in Chinese), 56(5): 658- 666.
Ren Z, Chen C, Pan K, et al. 2017. Gravity Anomalies of Arbitrary 3D Polyhedral Bodies with Horizontal and Vertical Mass Contrasts [J]. Surveys in Geophysics, 38(2): 479-502.
Suto N, Harada M, Izutsu, J et al. 2006. Time variation of the electromagnetic transfer function of the earth estimated by using wavelet transform[J]. Proc. Jpn. Acad., Ser. B, 82(5): 175-180.
Trad D O, Travassos J M. 2000. Wavelet filtering of magnetotelluric data [J]. Geophysics, 65(2): 482- 491.
WANG Yanwei, XIA Kewen, NIU Wenjia, et al. 2017. A denoising method by low-rank matrix recovery and its application in oil well logging [J]. Geophysical Prospecting for Petroleum (in Chinese), 56(5): 644- 650.
Weng Ai-hua, Li Zhou-bo, Mo Xiu-wen. 2003. Processing of low signal/ noise ratio NMR logging data [J]. Journal of Ji Lin university (earth science edition) (in Chinese), 33(2): 232-235.
Wu Lei, Kong Li, Cheng Jing-jing. 2011. Signal De-noising Algorithm Design in NMR Logging Based on Wavelet Transform [J]. Instrument Technique and Sensor (in Chinese), 10: 71-73, 83.
Xiao Li-zhi, Xie Ran-hong. 2003. Applications of NMR to oil well logging and formation evaluation [J]. Engineering Science (in Chinese), 5(9): 87-94.
蔡剑华, 肖晓. 2015. 基于小波自适应阈值去噪的MT信号处理方法[J]. 地球物理学进展, 30(6): 2433-2439, doi: 10.6038/pg20150601.
郭继亮, 李宏兵, 李明, 等. 2016. 一种体现孔隙形态影响的四参数孔隙度反演方法[J]. 石油物探, 55(4): 576-586.
罗新刚, 王万银. 2017. 余弦变换和Fourier变换在位场数据处理与转换中的对比研究[J]. 地球物理学进展, 32(3): 1197-1207, doi: 10.6038/pg20170333.
秦晅, 蔡建超, 刘少勇, 等. 2017. 基于经验模态分解互信息熵与同步压缩变换的微地震信号去噪方法研究[J]. 石油物探, 56(5): 658- 666.
王艳伟, 夏克文, 牛文佳, 等. 2017. 基于低秩矩阵恢复的去噪方法在石油测井中的应用[J]. 石油物探, 56(5): 644- 650.
翁爱华, 李舟波, 莫修文. 2003. 低信噪比核磁共振测井资料的处理技术[J]. 吉林大学学报(地球科学版), 33(2): 232-235.
吴磊, 孔力, 程晶晶. 2011. 基于小波变换的核磁共振测井信号去噪算法设计[J]. 仪表技术与传感器, 10: 71-73, 83.
肖立志, 谢然红. 2003. 核磁共振在石油测井与地层油气评价中的应用[J]. 中国工程科学, 5(9): 87-94.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |