Full waveform inversion based on principal component analysis and gradient reconstruction
Shi Caiwang1,2,3, He Bingshou1,2,3
1. Ocean University of China, Qingdao, Shandong 266100, China; 2. Laboratory of Marine Mineral Resources Evaluation and Detection, Qingdao National Laboratory for Marine Science and Technology, Qingdao, Shandong 266071, China; 3. Key Laboratory of Submarine Geosciences and Prospecting Technology, Ministry of Education, Qingdao, Shandong 266100, China
Abstract:When applied to heavily noisy data,full waveform inversion (FWI) usually provides a terrible result because conventional FWI concentrates on complete consistency between simulated data and original records.This article analyzes the influence of random noise on FWI and proposes a new gradient processing method based on principal component analysis (PCA) and gradient reconstruction.Firstly we apply PCA to the matrix consisted of every shot gradient,and pick specific principal components with high representativeness.Then we can reconstruct gradient with these principal components.When the signal-to-noise ratio (SNR) of residuals is relatively high,this method can reconstruct accurate gradient which will help the objective function decrease effectively.If the SNR is low,the reconstructed gradient will prevent the inversion process from accessing to a wrong model,which can provide a more reasonable initial model for the next frequency band.Numerical experiments show that FWI based on PCA and gradient reconstruction is more robust than conventional methods,which obtains acceptable results even when the SNR is low.
史才旺, 何兵寿. 基于主成分分析和梯度重构的全波形反演[J]. 石油地球物理勘探, 2018, 53(1): 95-104.
Shi Caiwang, He Bingshou. Full waveform inversion based on principal component analysis and gradient reconstruction. Oil Geophysical Prospecting, 2018, 53(1): 95-104.
Tarantola A.Inversion of seismic data in acoustic approximation.Geophysics,1984,49(8):1259-1266.
[2]
Bunks C,Saleck F M,Zaleski S et al.Multiscale seismic waveform inversion.Geophysics,1995,60(5):1457-1473.
[3]
Pratt R G.Seismic waveform inversion in the frequency domain,Part 1:Theory and verification in a physical scale model.Geophysics,1999,64(3):888-901.
[4]
Shin C,Cha Y H.Waveform inversion in the Laplace domain.Geophysical Journal International,2008,173(3):922-931.
[5]
Sirgue L,Etgen J T,Albertin U.3D frequency domain waveform inversion using time domain finite difference methods.EAGE 70th Conference and Technical Exhibition,2008.
[6]
Bekara M,Van der Baan M.Local SVD/ICA for signal enhancement of pre-stack seismic data.EAGE 68th Conference and Technical Exhibition,2006.
[7]
沈鸿雁,李庆春.频域奇异值分解(SVD)地震波场去噪.石油地球物理勘探,2010,45(2):185-189.Shen Hongyan,Li Qingchun.SVD(singular value decomposition) seismic wave field noise elimination in frequency domain.OGP,2010,45(2):185-189.
[8]
高静怀,毛剑,满蔚仕等.叠前地震资料噪声衰减的小波域方法研究.地球物理学报,2006,49(4):1155-1163.Gao Jinghuai,Mao Jian,Man Weishi et al.On the denoising method of prestack seismic data in wavelet domain.Chinese Journal of Geophysics,2006,49(4):1155-1163.
[9]
张恒磊,刘天佑,张云翠.基于高阶相关的Curvelet域和空间域的倾角扫描噪声压制方法.石油地球物理勘探,2010,45(2):208-214.Zhang Henglei,Liu Tianyou,Zhang Yuncui.High order correlation based dip angle scanning noise elimination method in Curvelet domain and space domain.OGP,2010,45(2):208-214.
[10]
张华,陈小宏,李红星等.曲波变换三维地震数据去噪技术.石油地球物理勘探,2017,52(2):226-232.Zhang Hua,Chen Xiaohong,Li Hongxing et al.3D seismic data de-noising approach based on Curvelet transform.OGP,2017,52(2):226-232.
[11]
Tarantola A.Inverse Problem Theory:Methods for Data Fitting and Model Parameter Estimation.Elsevier,Amsterdam and New York,1987.
[12]
Hugues A D,Tarantola A.Multiparameter L1 norm waveform fitting:interpretation of Gulf of Mexico reflection seismograms.Geophysics,1999,64(4):1023-1035.
[13]
Ha T,Chung W,Shin C.Waveform inversion using a back-propagation algorithm and a Huber function norm.Geophysics,2009,74(3):R15-R24.
[14]
Bube K P.Hybrid L1/L2 minimization with applica-tions to tomography.Geophysics,1997,62(4):1183-1195.
[15]
Brossier R,Operto S,Virieux J.Robust elastic frequency-domain full-waveform inversion using the L1 norm.Geophysical Research Letters,2009,36(20):L20310.
[16]
成景旺,吕晓春,顾汉明等.基于柯西分布的频率域全波形反演.石油地球物理勘探,2014,49(5):940-945.Cheng Jingwang,Lü Xiaochun,Gu Hanming et al.Full waveform inversion with Cauchy distribution in the frequency domain.OGP,2014,49(5):940-945.
[17]
Oh J W,Min D J.Robust scaling strategy for frequency-domain acoustic full waveform inversion.ASEG 23rd Geophysical Conference Extended Abstracts,2013,1-4.
[18]
Brossier R,Operto S,Virieux J.Which data residual norm for robust elastic frequency-domain full waveform inversion?.Geophysics,2010,75(3):R37-R46.
[19]
Oh J W,Min D J.Spectral filtering of gradient for L2-norm frequency-domain elastic waveform inversion.Geophysical Journal International,2013,193(2):820-840.
[20]
Kwon T,Seol S J,Byun J.Reduction of noise effect on elastic full-waveform inversion by using plane-wave data.SEG Technical Program Expanded Abstracts,2016,35:1131-1135.
[21]
苗永康.基于L-BFGS算法的时间域全波形反演.石油地球物理勘探,2015,50(3):469-474.Miao Yongkang.Full waveform inversion in time domain based on limited-memory BFGS algorithm.OGP,2015,50(3):469-474.
[22]
Shin C,Jang S,Min D J.Improved amplitude preservation for prestack depth migration by inverse.Geophysical Prospecting,2008,49(5):592-606.
[23]
Ben-Hadj-Ali H,Operto S,Virieux J.An efficient frequency-domain full waveform inversion method using simultaneous encoded sources.Geophysics,2011,76(4):R109-R124.
[24]
王学民.应用多元分析.上海:上海财经大学出版社,2014,251-256.
[25]
Liu C,Han M,Han L et al.Application of principal component analysis for frequency-domain full waveform inversion.SEG Technical Program Expanded Abstracts,2012,31:1-5.
[26]
段超然,韩立国.主成分分析波场重构反演与全波形反演联合速度重构.石油地球物理勘探,2016,51(6):1134-1140.Duan Chaoran,Han Liguo.A joint velocity reconstruction method:principal component analysis based wavefield-reconstruction inversion combined with full waveform inversion.OGP,2016,51(6):1134-1140.
[27]
王凌群,李冰冰,林君等.航空电磁数据主成分滤波重构的噪声去除方法.地球物理学报,2015,58(8):2803-2811.Wang Lingqun,Li Bingbing,Lin Jun et al.Noise removal based on reconstruction of filtered principal components.Chinese Journal of Geophysics,2015,58(8):2803-2811.
[28]
Zhu K G,Ma M Y,Che H W et al.PC-based artificial neural network inversion for airborne time-domain electromagnetic data.Applied Geophysics,2012,9(1):1-8.