摘要
随着CPU/GPU机群性能的提高及广泛应用,近几年来逆时偏移得到迅速发展完善并已在国内外投入大规模生产.由于逆时偏移采用双程波波动方程,在逆掩断层、推覆体等复杂地质构造成像方面明显优于单程波偏移方法.本文尝试将三角网格算法引入逆时偏移中,使其可直接处理起伏地表采集的数据,而无需预先做静校正.基于三角网格的有限差分算法保持了差分算法的简单性和有限元算法的精确性.与"波场下延累加"法单程波偏移不同的是,基于矩形网格的逆时偏移算法在直接处理复杂地表数据时,极易出现不稳定的情况,难以实用.而基于三角网格的逆时偏移算法能够保证波场外推过程的稳定性,并很好地使近地表漫散射收敛.本文通过对模型数据和实际资料的测试验证了该方法的有效性.
With the intensive application of high-performance CPU/GPU computer clusters,reverse-time migration has been developed well and implemented in exploration industry during recent years.Due to adoption of double-wave equation,reverse-time migration has significant advantage over other single-wave migration methods,especially in imaging for area such as overthrust,deeply dipping faults and so on.It is highly accurate and not affected by strong subsurface structure complexity and horizontal velocity variations.The difference method based on triangular grids maintains the simplicity of the difference method and the precision of the finite element method.It can be used directly for forward modeling on models with complex top surfaces and migration without statics preprocessing.Different from ＂wavefield extrapolation and accumulation＂ single-wave migration,rectangle grid based reverse-time migration may easily runs into instability when processing data from area with irregular top surface,which is not applicable in practice.However,triangular grid based RTM can ensure the stability of wave continuation and collapse diffraction waves at near surface perfectly.We have applied a finite difference method based on triangular grids for pre-stack reverse-time migration.Tests on model data and field data verify that the method introduced here can achieve perfect results in application.
出处
《地球物理学进展》
CSCD
北大核心
2012年第5期2077-2083,共7页
Progress in Geophysics
基金
国家"973"计划(2007CB209605)
国家自然科学基金(40974073)
山东省优秀中青年科学家科研奖励基金(BS2010NJ022)联合资助
关键词
逆时偏移
有限差分
三角网格
起伏地表
reverse-time migration
finite difference
triangular grid
complex top surface
作者简介
孙小东，男（汉族），1980年生，中国石油大学讲师、博士，主要从事地震成像和地震波正演模拟方面的科研工作．（E-mail：wanliliuyun@163．com）
