三角洲砂岩尖灭线的地震匹配追踪瞬时谱识别方法
|
摘要
本文在介绍地震信号匹配追踪的基本原理和高精度瞬时谱的计算方法的基础上。通过分析楔形尖灭体的时域和频域响应特征,提出了一种在时频域利用瞬时谱特征识别砂岩尖灭位置的新方法。针对三角洲构造的不同方向的尖灭厚度是变化的,本文提出利用核主成分分析方法,将优势调谐频带内各瞬时谱分量包含的尖灭线有效信息进行综合,即可展示出尖灭线的全貌,从而减少了三角洲尖灭线解释的不确定性。
This paper first presents the principle of seismic signal matching pursuit decomposition and the computation of high precision instantaneous spectra,then proposes a new method for wedge edge interpretation with its instantaneous spectral features in time-frequency domain by analyzing the time domain and frequency domain responses of the wedge edge.Actually,the delta fringe deposit thickness is not the same along different direction.In this situation,this paper introduces kernel principal component analysis method to integrate the effective information of pinch-out line contained in the original dominant instantaneous spectral components.Therefore,the whole face of pinch-out line can be fully recognized and the non-uniqueness in delta fringe interpretation is extremely reduced.
This paper first presents the principle of seismic signal matching pursuit decomposition and the computation of high precision instantaneous spectra,then proposes a new method for wedge edge interpretation with its instantaneous spectral features in time-frequency domain by analyzing the time domain and frequency domain responses of the wedge edge.Actually,the delta fringe deposit thickness is not the same along different direction.In this situation,this paper introduces kernel principal component analysis method to integrate the effective information of pinch-out line contained in the original dominant instantaneous spectral components.Therefore,the whole face of pinch-out line can be fully recognized and the non-uniqueness in delta fringe interpretation is extremely reduced.
引文
[1]Castagna J P,Sun Shengjie et al.Instantaneous spec-tral analysis:Detection of low-frequency shadows as-sociated with hydrocarbons.The Leading Edge,2003,22(2):120～127
[2]Castagna J P,Sun Shengjie.Comparison of spectraldecomposition methods.First Break,2006,24(1):75～79
[3]Chuang H,Lawton D C.Amplitude responses of thinbeds:sinusoidal approximation versus Ricker approxi-mation.Geophysics,1995,60(1):223～230
[4]Chuang H,Lawton D C.Frequency characteristics ofseismic reflections from thin beds.Canadian Journalof Exploration Geophysics,1995,31(1):32～37
[5]Carlos A,Johan A K.Multiway spectral clusteringwith out-of-sample extensions through weighted ker-nel PCA.IEEE Transactions on Pattern Analysisand Machine Intelligence,2010,32(2):335～347
[6]Mallat S,Zhang Z.Matching pursuit with time-fre-quency dictionaries.IEEE Transactions on SignalProcessing,1993,41(12):3397～3415
[7]张繁昌,李传辉,印兴耀.基于动态匹配子波库的地震数据快速匹配追踪.石油地球物理勘探,2010,45(5):667～673Zhang Fanchang,Li Chuanhui and Yin Xingyao.Seis-mic data fast matching pursuit based on dynamicmatching wavelet library.OGP,2010,45(5):667～673
[8]Liu J,Marfurt K J.Instantaneous spectral attributesto detect channels.Geophysics,2007,72(1):23～31
[9]Liu J,Wu Y,Han D et al.Time-frequency decompo-sition based on Ricker wavelet.SEGTechnicalProgramExpandedAbstracts,2004,23:1937～1940
[10]Widess M B.How thin is a thin bed?Geophysics,1973,38(6):1176～1180
[11]Sun Luping,Zheng Xiaodong.Thin-bed thickness cal-culation formula and its approximation using peak fre-quency.Applied Geophysics,2009,6(3):234～240
[12]Tai S H,Puryear C,Castagna J P.Local frequency asa direct hydrocarbon indicator.SEG Technical Pro-gram Expanded Abstracts,2009,28:2160～2164
[2]Castagna J P,Sun Shengjie.Comparison of spectraldecomposition methods.First Break,2006,24(1):75～79
[3]Chuang H,Lawton D C.Amplitude responses of thinbeds:sinusoidal approximation versus Ricker approxi-mation.Geophysics,1995,60(1):223～230
[4]Chuang H,Lawton D C.Frequency characteristics ofseismic reflections from thin beds.Canadian Journalof Exploration Geophysics,1995,31(1):32～37
[5]Carlos A,Johan A K.Multiway spectral clusteringwith out-of-sample extensions through weighted ker-nel PCA.IEEE Transactions on Pattern Analysisand Machine Intelligence,2010,32(2):335～347
[6]Mallat S,Zhang Z.Matching pursuit with time-fre-quency dictionaries.IEEE Transactions on SignalProcessing,1993,41(12):3397～3415
[7]张繁昌,李传辉,印兴耀.基于动态匹配子波库的地震数据快速匹配追踪.石油地球物理勘探,2010,45(5):667～673Zhang Fanchang,Li Chuanhui and Yin Xingyao.Seis-mic data fast matching pursuit based on dynamicmatching wavelet library.OGP,2010,45(5):667～673
[8]Liu J,Marfurt K J.Instantaneous spectral attributesto detect channels.Geophysics,2007,72(1):23～31
[9]Liu J,Wu Y,Han D et al.Time-frequency decompo-sition based on Ricker wavelet.SEGTechnicalProgramExpandedAbstracts,2004,23:1937～1940
[10]Widess M B.How thin is a thin bed?Geophysics,1973,38(6):1176～1180
[11]Sun Luping,Zheng Xiaodong.Thin-bed thickness cal-culation formula and its approximation using peak fre-quency.Applied Geophysics,2009,6(3):234～240
[12]Tai S H,Puryear C,Castagna J P.Local frequency asa direct hydrocarbon indicator.SEG Technical Pro-gram Expanded Abstracts,2009,28:2160～2164
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心