The Algorithm of High Order Finite Difference Pre‐Stack Reverse Time Migration and GPU Implementation

et al. 2011

Self Cite

Pre-stack reverse time migration (RTM) for rugged topography is a very useful tool for seismic imaging in the areas with strong relief on the surface and complex subsurface structures. In this paper, we illustrate the implementation process of RTM for rugged topography. To deal with the difficulty of dealing with the rugged free boundary condition with the finite difference method, we employ a simplified boundary condition which could avoid the abundant logical judgment. On this basis, we use the Graphic Processing Unit (GPU) architecture to accelerate RTM and get an order of magnitude higher speedup ratio compared to the traditional CPU architecture. The tests on synthetic data examples and comparison with the pre-stack one-way wave method for rugged topography prove that RTM does not have the imaging dip limit, and the imaging results for the areas with rugged surface topography and subsurface steep structures are significantly improved. The problems of imaging noise removal and massy memory demand of RTM have been stated in our previous paper and will not be discussed here.

Section: Reverse Time Migration From Rugged Topographymentioning

confidence: 99%

“…Micikevicius [32] discussed the GPU realization process of the high-order finite difference method. Liu [33,34] and Foltinek [35] use GPU to realize RTM. In this article, we will mainly discuss the process of RTM from rugged topography and GPU acceleration.…”

Section: Introductionmentioning

confidence: 99%

Pre‐Stack Reverse Time Migration for Rugged Topography and GPU Acceleration Technology

et al. 2011

Self Cite

“…In China, Di et al [16] studied RTM by FE on elastic waves. Zhang et al [17] , He et al [18−19] , Du et al [20] , and Liu et al [21] conducted RTM using highorder FD on scalar and vector wave fields separately. Wang et al [22] attempted to cope with this task by the spectral element method.…”

Section: Introductionmentioning

confidence: 99%

Acoustic Prestack Reverse Time Migration Using the Adaptive High‐Order Finite‐Difference Method in the Time‐Space Domain

Yan

2013

At present, prestack reverse time migration (RTM) is the most accurate method for seismic imaging in theory. One of its key steps is wavefield extrapolation. The essence of wavefield extrapolation is to solve the wave equation. Thus solving the wave equation fast yet accurately is very important. In this paper, we use a time‐space domain dispersion‐relation‐based finite‐difference (FD) method to solve the acoustic wave equation, then analyze the dispersion and stability of the numerical solution, perform the numerical modeling, and compare the results with the conventional FD method. The results of analysis and numerical modeling demonstrate the time‐space domain FD method has better accuracy and stability than the conventional method under the same discretization. We adopt the time‐space domain high‐order FD method to solve the acoustic wave equation in the process of wavefield extrapolation, then use the normalized cross‐correlation imaging condition to obtain imaging in RTM. The results of synthetic data processing show that the imaging accuracy has enhanced. Meanwhile, in the process of wavefield extrapolation, we adopt adaptive variable‐length spatial operators to compute spatial derivatives and thus to decrease computing cost significantly without reducing the accuracy of numerical modeling and imaging.

“…Du [19] applied the pre-stack RTM to multi-component data. Liu [20] realized high-order finite difference RTM using GPU with a speedup ratio about 30.…”

Section: Introductionmentioning

confidence: 99%

“…Highly intensive computation cost, low-frequency imaging noise and massy memory demand are bottlenecks of pre-stack RTM. The problem of computation cost has been addressed in our previous work [20] and this paper focuses on the other two problems.…”

Section: Introductionmentioning

confidence: 99%

Denoising and Storage in Seismic Reverse Time Migration

Zou

et al. 2010

Pre‐stack reverse time migration (RTM) is a very useful tool for seismic imaging. It has, however, some problems such as highly intensive computation cost, low‐frequency imaging noise and massy memory demand. The problem of time consuming for calculation can be solved by GPU/CPU collaborative computation. This paper focuses on suppressing noise and data storage. First, we analyze the generation mechanism of low‐frequency imaging noise and suggest that the phase and frequency spectra of seismic data are modified before migration and a Laplacian filter is used to remove the low wavenumber noise efficiently. Aiming at the problem of massy memory demand, we adopt the random boundary condition which sacrifices the computation cost but reduces the memory demand. The implementation can be performed on GPU and the communication between CPU and GPU can be saved to reduce the computation cost in another way. The tests on synthetic data examples illustrate that the difference between the migration result from the method of random boundary condition and that from the traditional method of saving the footprint of the wavefield can be overlooked.