基于GPU加速的包络波形反演  

GPU-based envelop waveform inversion

作　　者：李振春[1] 司道军 孙思宇 张凯[1] LI Zhenchun SI Daojun SUN Siyu ZHANG Kai(School of Geosciences, China University of Petroleum （East China） , Qingdao 266580,Chin)

机构地区：[1]中国石油大学(华东)地球科学与技术学院,青岛266580

出　　处：《物探化探计算技术》2017年第2期231-236,共6页Computing Techniques For Geophysical and Geochemical Exploration

基　　金：国家自然科学基金(41204086);国家科技重大专项课题(2011ZX05006-002)

摘　　要：波形反演利用地震资料的相位和振幅信息对速度进行建模,是当前分辨率最高的速度估计方法。波形反演对初始速度场具有较高的要求,当初始速度场和真实的速度场相差较远时,容易产生周波跳跃问题。通常在进行速度反演时,采用多尺度方法减轻波形反演的非线性,然而当实际地震资料中缺少大偏移距和低频数据时,多尺度方法不再适用。此外,计算效率也限制了波形反演的应用。因此,这里提出基于GPU加速的包络波形反演来获取包含长波长信息的初始速度场,并提高计算效率。通过对Marmousi模型的测试,验证了方法的正确性和有效性。Waveform inversion utilizing phases and amplitudes of seismic data to construct velocity model is the velocity es-timation method with the highest resolution. However, it requires relatively accurate initial velocity. When the initial velocity is far from the true velocity, circle-skipping problem arises. Multi-scale method is often used to mitigate the nonlinearity of waveform inversion, but it fails when seismic data dont have long - offset and low - frequency information. In addition, compu-tational efficiency restricts the application of waveform. Thus？ we propose GPU -based envelop waveform inversion in this pa-per to obtain initial velocity field with long wavelength information and improve computational efficiency. And test on Marmou- si model verifies the correctness and validity of the method proposed in this paper.

关 键 词：波形反演 包络 GPU加速 速度估计 周波跳跃 

分 类 号：P631.4[天文地球—地质矿产勘探]