基于稀疏贝叶斯的分频蚂蚁追踪技术在裂缝刻画中的应用  

Application of frequency division ant tracking technology based on Sparse Bayesian in fracture characterization

作　　者：韦豪 王元君[1,2] 周怀来[1,2,3] 王肖 WEI Hao;WANG Yuanjun;ZHOU Huailai;WANG Xiao(Chengdu University of Technology College of Geophysics,Chengdu 610059,China;Chengdu University of TechnologyState Key Laboratory of Oil and Gas Reservoir Geology and Exploitation,Chengdu 610059,China;Chengdu University of Technology Key Laboratory of Earth Exploration and Information Technology of Ministry of Education,Chengdu 610059,China)

机构地区：[1]成都理工大学地球物理学院,成都610059 [2]成都理工大学油气藏地质及开发工程国家重点实验室,成都610059 [3]成都理工大学地球探测与信息技术教育部重点实验室,成都610059

出　　处：《物探化探计算技术》2023年第2期192-201,共10页Computing Techniques For Geophysical and Geochemical Exploration

基　　金：四川省科技计划重点研发项目(2021YFG0257);国家自然科学基金(42204138)。

摘　　要：在裂缝预测方法中,分频处理是常用的裂缝预测方法之一,这对时频分析方法的抗噪性和分辨率都有很高的要求。这里用稀疏贝叶斯方法(Sparse Bayesian Learning,SBL)对地震信号进行子波重构,该方法在地震模型设置时,将噪声模型考虑其中,再用威格纳分布(Wigner-ville Distribution,WVD)进而得到地震信号的时频分布,称之为稀疏贝叶斯时频分析方法(SBL-WVD)。该方法不仅有很好的抗噪性,还拥有较高的时频分辨率。考虑到高频相干对中小尺度裂缝带预测效果较好,而蚂蚁追踪技术对裂缝地刻画更清晰,因此采用分频相干以及蚂蚁追踪这种综合方法对裂缝进行预测。应用到绥中某工区发现:该方法对中小尺度裂缝刻画清晰,在实际应用中取得良好的效果,为后期的储层识别提供借鉴。In the fracture prediction methods frequency division processing is one of the commonly used fracture prediction methods,which has high requirements for the noise resistance and resolution of time-frequency analysis methods.This paper uses sparse Bayesian learning(SBL)to reconstruct the wavelet of seismic signals.When setting the seismic model,the noise model is considered,and then the time-frequency distribution of the seismic signal is obtained by Wigner Ville distribution(WVD).This paper is called sparse Bayesian time-frequency analysis method(sbl-wvd).This method has not only good noise resistance and high time-frequency resolution.High-frequency coherence has a better prediction effect on small and medium-sized fracture zones.Furthermore,ant-tracking technology has a more precise description of fractures.Therefore,the comprehensive frequency division coherence and ant trackingis used to predict fractures.When applied to a work area in Suizhong,it is found that this method can predict small and medium-sized fractures.The fracture description is clear,which has achieved good results in practical application and laid a foundation for reservoir identification in the later stage.

关 键 词：稀疏贝叶斯时频分析方法(SBL-WVD) 倾角导向相干 蚂蚁追踪技术 裂缝预测 

分 类 号：P631.4[天文地球—地质矿产勘探]