海量数据叠前时间偏移方法

THE METHOD OF PRESTACK TIME MIGRATION IN PROCESSING ENORMOUS DATA

在四川盆地的礁滩天然气勘探中,三维地震勘探控制面积近4000km2,满覆盖面积约2500km2,处理数据达36259.44万道,总数据量近7TB。如何高质量、快速完成叠前时间偏移是一个非常困难的课题。为此,在偏移参数试验和高精度速度分析之后,经过单作业节点数的偏移时间和多作业节点组的偏移时间测试,选取最佳偏移节点与作业数组合,采用无缝拼接方法,高效优质地完成了大数据量三维叠前时间偏移。该方法为地震数据的处理提供了一种新手段。

During natural gas exploration in reef flat of the Sichuan basin,there are 4000 km2 area of 3D seismic survey under control,2500 km2 among which is the domain with the data of 36 529.44 ten thousand seismic traces(that is to say,7×1012 bytes of the data in total) to be processed.Obviously,it is difficult for the operators to get a reliable image from the prestack migration of 3D land seismic surveys because of such a large number of data.Therefore,this paper introduces a method of prestack time migration in dealing with such a mass of data.Based on thorough analysis on high-fidelity velocity and migration parameters tests,this method carried out tests on the migration time of both nodes in a single case and nodes group in multiple cases,and realized seamless combination of the optimum nodes with the number of cases.This method provides a new means for engineers to process enormous seismic data.