节点文献

压缩三维走时表提高克希霍夫偏移计算效率

IMPROVING COMPUTING EFFICIENCY OF KIRCHHOFF MIGRATION BY COMPRESSING 3-D TRAVEL TIME TABLE

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 王珺杨长春李波涛

【Author】 WANG Jun1, YANG Chang-hun2, LI Bo-tao2 ( 1. University of Petroleum, Dong Ying 257061, China;2. Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China).

【机构】 石油大学信息与控制工程学院中科院地质与地球物理研究所中国科学院兰州地质研究所

【摘要】 地震波走时是克希霍夫叠前深度偏移的重要参数。三维克希霍夫偏移成像由于需频繁读取大数据量的走时表文件,所以计算效率不高。这里提出一种通过压缩三维射线追踪走时表,来提高克希霍夫偏移计算效率的方法:首先规划一个具有规则网格控制点并覆盖所有走时表点集的最小长方体区域,然后以三维三次B样条函数为插值基函数进行最小二乘法曲面体拟合,求出规则网格控制点的数值并以数组形式存储入内存,采用稀疏化存储进一步节省了内存空间。在偏移成像时,再由这些规则网格控制点的数值,使用线性插值公式解编出走时表。实际资料算例验证了该走时表压缩方法不仅近似精度高,计算稳定度高,计算效率高,而且由于省去了频繁进行大数据量走时表文件的读写操作,所以克希霍夫偏移的计算效率提高了二倍以上。

【Abstract】 Travel time of seismic waves is an important parameter of Kirchhoff migration. In process of 3-D Kirchhoff migration, the enormous travel time table file is frequently read, which leads to a fairly low calculation efficiency. A method is proposed to improve the computing efficiency of 3-D Kirchhoff migration by compressing the ray tracing traveltime table: At first,a smallest rectangle region in 3-D space with regular grids is constructed, where all the travel time data points are included. Then, the 3-D cubic B-spline function is used to fit a cuboidal surface which covers all the 3-D scattered travel time data, and the values of the finite regular grids can be calculated using least square method and are stored into memory in terms of an array. Compressed sparse column method is used to save more space. When the scattered travel time data are needed by migration, they can be decompressed by liner interpolation of 3-D B-spline function. Application to real seismic data shows that the traveltime compression method can not only compress travel time data with high approximate accuracy, stable calculation result, high calculation efficiency, but also improve the numerical efficiency of Kirchboff migration by twice because the enormous travel time table file neednt be frequently read any more.

【基金】 教育部高等教育博士点专项研究基金资助(200804251502);中国石油天然气集团公司创新基金资助(07E1019);中央高校基本科研业务费专项资金资助(09CX04013A)
  • 【文献出处】 物探化探计算技术 ,Computing Techniques for Geophysical and Geochemical Exploration , 编辑部邮箱 ,2011年02期
  • 【分类号】P631.44
  • 【下载频次】104
节点文献中: 

本文链接的文献网络图示:

本文的引文网络