基于匹配追踪算法子波分解技术在薄互层储层预测中的应用  

THE APPLICATION OF WAVELET DECOMPOSITION TECHNIQUE BASED ON MATCHING PURSUIT ALGORITHM IN THIN INTERBEDDED RESERVOIR PREDICTION

作　　者：何胡军[1] 王秋语[2] 程会明[2] 

机构地区：[1]成都理工大学能源学院,四川成都610059 [2]中国石化胜利油田地质科学研究院,山东东营257015

出　　处：《物探化探计算技术》2010年第6期641-644,568-569,共4页Computing Techniques For Geophysical and Geochemical Exploration

摘　　要：薄互层储层砂体预测技术的研究已成为国内、外各油田的重要研究课题,并已取得较多研究成果。但多数利用地震资料进行薄互层预测技术,均基于单一地震子波与地震反射系数函数的褶积假设,实际上地震子波在穿过不同物理特性地层时,其形状会发生不同变化,导致该假设与实际情况尚有一定距离,致使薄互层储层预测的精度降低。这里主要论述的是基于匹配追踪算法子波分解技术识别薄互层储层。通过研究表明,该技术能较好地分辨出储层在空间的展布规律,识别薄互层储层,为油藏的地质及地球物理解释提供有力的支持。The studies of prediction techniques for thin interbedded sand reservoir have become an important research topic and obtained many achievements.However,most of them using seismic data are based on convolution model,which is defined as the convolution of a single wavelet and a seismic reflection coefficient function.Actually,the shape of wavelet will change when seismic wavelets go through formation with different physical properties,therefore,this single wavelet convolution model is far away from the real situation,and it will decrease the accuracy of prediction.In this paper,we will discuss wavelet decomposition technique based on matching pursuit algorithm.The results show that this technique can identify the distribution of thin-interbedded sandstone reservoirs,and support for the reservoir geological and geophysical interpretation.

关 键 词：匹配追踪 子波 薄互层 分解与重构 地震数据体 

分 类 号：P631.443[天文地球—地质矿产勘探]