Background noise attenuation method of DAS seismic data based on multiscale enhanced cascade residual network
ZHONG Tie1,2, WANG Weiyu3, WANG Wei4, DONG Shiqi1,2, LU Shaoping5, DONG Xintong6
1. Key Laboratory of Modern Power System Simulation and Control and Renewable Energy Technology, Ministry of Education, Jilin, Jilin 132012, China; 2. College of Electrical Engineering, Northeast Electric Power University, Jilin, Jilin 132012, China; 3. Ziyang Electric Power Supply Company, State Grid Sichuan Electric Power Company, Ziyang, Sichuan 641300, China; 4. Research Institute of Petroleum Exploration & Development-Northwest, PetroChina, Lanzhou, Gansu 730020, China; 5. School of Earth Sciences and Engineering, SUN YAT-SEN University, Guangzhou, Guangdong 510275 China; 6. College of Instrumentation & Electrical Engineering, Jilin University, Changchun, Jilin 130026, China
Abstract:Seismic records collected through distributed optical fiber acoustic sensing (DAS) typically exhibit a low signal-to-noise ratio (SNR) due to the pervasive influence of complex and intense background noise. How to effectively suppress background noise,restore weak upgoing reflection information,and substantially improve the SNR of DAS records havs become a prominent challenge in seismic data processing. To address the issue of complex DAS background noise attenuation,this paper proposes a multiscale enhanced cascade residual network (MECRN),which employs a dual-path cascade residual network structure to extract shallow information from DAS records. On this basis,dilated convolutional layers and multiscale modules are introduced to extract the multiscale features existing in DAS records. Additionally,skip connections are introduced to import shallow features,which enhances the feature extraction capability of MECRN and avoids effective feature loss. Finally,the local and global features are integrated by residual learning,and the reconstructed features are refined to improve the denoising capabilities of MECRN. The processing results from both simulated and field DAS data demonstrate that MECRN can effectively suppresses complex DAS background noise and accurately restores weak reflection signals,which enhances the processing capacity of DAS data significantly.
基金资助:本项研究受国家自然科学青年基金项目“基于多尺度可迁移深度学习方法的多井DAS地震数据“智普”消噪技术研究”(42204114)、第6批博士后创新人才支持计划项目“基于对抗式深度学习策略的 DAS 地震资料智能消噪系统构建”(BX2021111)、吉林省科技厅面上基金项目“基于深度学习框架的复杂地震勘探资料智能消噪技术研究”(20220101190JC)、中国石油天然气集团公司前瞻性基础性项目“物探岩石物理与前沿储备技术研究”(2021DJ3505)及中国石油股份公司科技项目“川南页岩气开发区应力变化、构造活化与可能诱发地震机理研究”(2022DJ8004)联合资助。
蔡志东. 井中地震技术:连接多种油气勘探方法的桥梁[J]. 石油地球物理勘探,2021,56(4):922-934.CAI Zhidong. Borehole seismic:A bridge connecting multiple oil and gas exploration methods[J]. Oil Geophysical Prospecting,2021,56(4):922-934.
[2]
EGOROV A,CORREA J,BÓNA A,et al. Elastic full-waveform inversion of vertical seismic profile data acquired with distributed acoustic sensors[J]. Geophysics,2018,83( 3):R273-R281.
[3]
VERDON J P,HORNE S A,CLARKE A,et al. Microseismic monitoring using a fibre-optic distributed acoustic sensor (DAS) array[J]. Geophysics,2020,85(3):KS89-KS99.
[4]
林融冰,曾祥方,宋政宏,等. 分布式光纤声波传感系统在近地表成像中的应用Ⅱ:背景噪声成像[J]. 地球物理学报,2020,63(4):1622-1629.LIN Rongbing,ZENG Xiangfang,SONG Zhenghong,et al. Distributed acoustic sensing for imaging shallow structure Ⅱ:Ambient noise tomography [J]. Chinese Journal of Geophysics,2020,63(4):1622-1629.
[5]
DONG X,ZHONG T,LI Y. New suppression technology for low-frequency noise in desert region:The improved robust principal component analysis based on prediction of neural network[J]. IEEE Transactions on Geoscience and Remote Sensing,2020,58(7):4680-4690.
[6]
MENDEL J. White-noise estimators for seismic data processing in oil exploration[J]. IEEE Transactions on Automatic Control,1977,22(5):694-706.
[7]
CHEN K,SACCHI M D. Robust F-x projection filtering for simultaneous random and erratic seismic noise attenuation[J]. Geophysical Prospecting,2017,65(3):650-668.
[8]
CHEN Y,ZU S,WANG Y,et al. Deblending of simultaneous source data using a structure oriented space-varying median filter[J]. Geophysical Journal International,2020,222(3):1805-1823.
[9]
LIU G,CHEN X. Noncausal F-x-y regularized nonstationary prediction filtering for random noise attenuation on 3D seismic data[J]. Journal of Applied Geophysics,2013,93:60-66.
[10]
PINNEGAR C R,EATON D W. Application of the S transform to prestack noise attenuation filtering[J]. Journal of Geophysical Research,2003,108(B9):369-378.
[11]
TIAN Y A,LI Y,YANG B J. Variable-eccentricity hyperbolic-trace TFPF for seismic random noise attenuation[J]. IEEE Transactions on Geoscience and Remote Sensing,2014,52(10):6449-6458.
[12]
武国宁,于萌萌,王君仙,等. 应用平稳小波变换与深度残差网络压制地震随机噪声[J]. 石油地球物理勘探,2022,57(1):43-51.WU Guoning,YU Mengmeng,WANG Junxian,et al. Seismic random noise attenuation based on stationary wavelet transform and deep residual neural network[J]. Oil Geophysical Prospecting,2022,57(1):43-51.
[13]
ZHAO X,LI Y,ZHUANG G H,et al. 2-D TFPF based on contourlet transform for seismic random noise attenuation[J]. Journal of Applied Geophysics,2016,129:158-166.
[14]
DONG X T,JIANG H,ZHENG S,et al. Signal-to- noise ratio enhancement for 3C downhole microseismic data based on the 3D shearlet transform and improved backpropagation neural networks[J]. Geophysics,2019,84(4):V245-V254.
[15]
童思友,高航,刘锐. 基于Shearlet变换的自适应地震资料随机噪声压制[J]. 石油地球物理勘探,2019,54(4):744-750.TONG Siyou,GAO Hang,LIU Rui. Seismic random noise adaptive suppression based on the Shearlet transform[J].Oil Geophysical Prospecting,2019,54(4):744-750.
[16]
KUTYNIOK G,LEMVIG J,LIM W Q. Optimally sparse approximations of 3D functions by compactly supported Shearlet frames[J]. SIAM Journal on Mathematical Analysis,2012,44(4):2962-3017.
[17]
AMEZQUITA SANCHEZ J P,CHAVEZ ALEG- RIA O,VALTIERRA RODRIGUEZ M,et al. Detection of ULF geomagnetic anomalies associated to seismic activity using EMD method and fractal dimension theory[J]. IEEE Latin America Transactions,2017,15(2):197-205.
[18]
YU S,MA J. Complex variational mode decomposition for slop-preserving denoising[J]. IEEE Transactions on Geoscience and Remote Sensing,2018,56(1):586-597.
[19]
BUSTOS A,RUBIO H,CASTEJÓN C,et al. EMD-based methodology for the identification of a high-speed train running in a gear operating state[J]. Sensors,2018,18(3):793-800.
[20]
LIU X,CHEN X,LI J,et al. Nonlocal weighted robust principal component analysis for seismic noise attenuation[J]. IEEE Transactions on Geoscience and Remote Sensing,2021,59(2):1745-1756.
[21]
董烈乾,张慕刚,骆飞,等. 应用SVD约束的迭代反演混叠噪声压制方法[J]. 石油地球物理勘探,2021,56(1):57-61,117.DONG Lieqian,ZHANG Mugang,LUO Fei,et al. Blended noise suppression based on SVD constrained iterative inversion[J]. Oil Geophysical Prospecting,2021,56(1):57-61,117.
[22]
严哲,顾汉明,蔡成国. 基于各向异性扩散滤波的地震图像增强处理[J]. 石油地球物理勘探,2013,48(3):390-394.YAN Zhe,GU Hanming,CAI Chengguo. Seismic image enhancement based on anisotropic diffusion[J]. Oil Geophysical Prospecting,2013,48(3):390-394.
[23]
YU G,CAI Z,CHEN Y,et al. Borehole seismic survey using multimode optical fibers in a hybrid wireline[J]. Measurement,2018,125:694-703.
[24]
SOTO M A,RAMÍREZ J A,THÉVENAZ L. Intensifying the response of distributed optical fibre sensors using 2D and 3D image restoration[J]. Nature Communications,2016,7(1):10870.
[25]
MARTINS H F,FERNÁNDEZ-RUIZ M R,COSTA L,et al. Monitoring of remote seismic events in metropolitan area fibers using distributed acoustic sensing (DAS) and spatiotemporal signal processing[C]. Optical Fiber Communication Conference and Exhibition (OFC),2019,1-3.
[26]
HINTON G E,OSINDERO S,TEH Y W. A fast learning algorithm for deep belief nets[J]. Neural Computation,2006,18(7):1527-1554.
[27]
ARUN P V,BUDDHIRAJU K M,PORWAL A,et al.CNN-based super-resolution of hyperspectral ima-ges[J]. IEEE Transactions on Geoscience and Remote Sensing,2020,58(9):6106-6121.
[28]
YU D,HINTON G,MORGAN N,et al. Introduction to the special section on deep learning for speech and language processing[J]. IEEE Transactions on Audio,Speech and Language Processing,2012,20(1):4-6.
[29]
JING Y,YANG Y,FENG Z,et al. Neural style transfer:A review[J]. IEEE Transactions on Visualization and Computer Graphics,2020,26(11):3365-3385.
[30]
LECUN Y,BENGIO Y,HINTON G. Deep learning[J]. Nature,2015,521(7553):436-444.
[31]
ZHANG K,ZUO W M,CHEN Y J,et al. Beyond a Gaussian denoiser:Residual learning of deep CNN for image denoising[J]. IEEE Transactions on Image Processing,2017,26(7):3142-3155.
[32]
MA H,YAO H,LI Y,et al. Deep residual encoder-decoder networks for desert seismic noise suppression[J]. IEEE Geoscience and Remote Sensing Letters,2020,17(3):529-533.
[33]
DONG X,LI Y. Denoising the optical fiber seismic data by using convolutional adversarial network based on loss balance[J]. IEEE Transactions on Geoscience and Remote Sensing,2021,59(12):10544-10554.
[34]
FENG Q,LI Y. Denoising deep learning network based on singular spectrum analysis:DAS seismic data denoising with multichannel SVDDCNN[J]. IEEE Transactions on Geoscience and Remote Sensing,2022,60:5902911.
[35]
VAN DEN ENDE M,LIOR I,AMPUERO J P,et al. A self-supervised deep learning approach for blind denoising and waveform coherence enhancement in distributed acoustic sensing data[J]. IEEE Transaction on Neural Networks and Learning Systems,2023,34(7):3371-3384.
[36]
ZHAO Y X,LI Y,WU N. Distributed acoustic sensing vertical seismic profile data denoiser based on convolutional neural network[J]. IEEE Transactions on Geoscience and Remote Sensing,2022,60:1-11.
[37]
董新桐,钟铁,王洪洲,等. 基于卷积对抗降噪网络的塔里木盆地沙漠地震资料消噪方法研究[J]. 地球物理学报,2022,65(7):2661-2672.DONG Xintong,ZHONG Tie,WANG Hongzhou,et al. The denoising of desert seismic data acquired from Tarim basin based on convolutional adversarial denoising network [J]. Chinese Journal of Geophysics,2022,65(7):2661-2672.
[38]
ZHAO Y,LI Y,DONG X,et al. Low-frequency noise suppression method based on improved DnCNN in desert seismic data[J]. IEEE Geoscience and Remote Sensing Letters,2019,16(5):811-815.