文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > 工程科技 > 能源/化工 > Near Surface Data Constrained Velocity Model Tomographic Inversion Technique
侵权投诉

Near Surface Data Constrained Velocity Model Tomographic Inversion Technique

2013, NAGAPC, 北京, 亚太近地表会议

Zhao Shengtian, Diao Rui *

Geophysical Research Institute of Shengli Oil Field Branch, Sinopec

Summary

In typical alluvial plain of maturing oil field in eastern China, the near surface structure can divide into lower velocity layer and subweathered layer. The distinguishing features are low velocity and loose textured structure. The velocity of near surface is changed rapidly in vertical and lateral direction. Usually the first break of seismic data is used for near surface velocity tomographic inversion in processing. This method is introduced to describe spatial velocity variation. Due to the restriction of acquisition data, the inversion model is usually different from low velocity layers (LVL) investigation result. The measured near surface data has high accuracy in control point but poor contingency & limited spatial distribution.

The near surface constrained tomographic inversion method can overcome the defect of above two methods. First, add the measured near surface data into initial velocity model through tomography grid interpolation method. And then, iterate the updated initial velocity model by first break tomographic inversion. The tomographic inversion method based on measured LVL data can obtain accurate near surface velocity model, which can achieve the reconstruction of near surface velocity model effectively.

Introduction

The near surface structure of maturing field in Shengli oil field is influenced by diversion of river. The near surface is flat, but lithologic character of sediments is complex alternation, which results in the complex variation of near surface thickness and velocity. The low velocity layer can introduce obvious delay times in recorded seismic data.

Refraction method is the basis theory of static correction in previous seismic exploration, such as: short refraction, routine refraction static correction and residual static correction of first break [1]. The assumption of refraction method is that layer medium constitute geologic model, and refracting surface is sustained [2]. But real surface is rough, and lateral velocity is change rapidly. Therefore, we introduce the measured LVL data constrained tomographic inversion to resolve static correction problems.

1 Major Theory

Tomographic inversion based on first break can obtain velocity model of near surface, which can calculate static

Near Surface Data Constrained Velocity Model Tomographic Inversion Technique

correction accurately. The influence of LVL and topography variation can be eliminated successfully. The method assumes the first break is collection of turning wave of first arrivals. Its nonlinear feature is suitable for complicated medium. The near surface velocity model can be achieved by vertical velocity gradient inversion [3].

The first break tomographic inversion method has two difficult problems in field data processing. One is massive data introduce first break picking efficiency problem, especially for low signal noise ratio data. The other problem is that near surface velocity model difference between inversion result and real measured data. The near surface velocity of tomographic inversion is higher than measured velocity, and near surface layer of tomography inversion is thicker than measured result. All of those cannot meet the demand of high precision processing.

So, we put forward a near surface data constrained tomographic inversion method for precise near surface model calculation. The inversion method includes two key points. (1) First break auto picking technique. In this method, the digital image processing technique such as boundary detection and boundary tracing is introduced to pick up the first break. The technique can guarantee the accuracy and continuity of picking result. (2) LVL data constrained tomographic inversion. Low velocity layer is measured by short refraction and micro-logging. And the LVL velocity keep tinily changes during the iterations of tomographic inversion. Therefore, the near surface velocity of inversion result depends on the measured data. This method compensates accurate velocity information of near surface zone which is lost in routine first break tomographic inversion.

2 First break auto picking technique

2.1 Seismic data normalization processing

The seismic traces have a wide range of amplitude value, which will influence the picking results seriously without preconditioning. Therefore, the normalization work should be implemented before picking.

2.2 Gray processing of seismic data

The gray processing of seismic data can be considered as a process that calculates absolute value of amplitude for each traces and samples. First, image the data according to the amplitude polarity. Second, reverse the negative amplitude. Finally, convert the seismic data into gray image for

第1页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email