| 167 | 0 | 38 |
| 下载次数 | 被引频次 | 阅读次数 |
地面微震监测因施工简便、成本低而广泛应用，但其信号易受噪声干扰。本文提出一种基于时-频(time frequency, TF)域深度学习的方法，通过同步挤压变换构建U-Net训练集，结合迁移学习策略消除低信噪比数据中的噪声。实际监测数据所包含的有效微震信号数量有限，难以被用来构建足够大的标签训练集。本文采用先使用模拟数据集进行训练，再使用实际数据进行迁移学习的策略。首先，对无噪的模拟微震信号和加入真实噪声的模拟微震信号分别进行同步挤压变换，获取它们的高分辨率时-频谱，并用其构建U-Net网络的时-频域训练集。网络训练完成后，再使用实际微震监测信号所构成的小规模训练集对网络进行迁移学习，并得到最终的消噪网络。使用多道微震监测数据的高分辨率时-频谱构建训练集，可以更好地利用监测数据道与道之间有效微震信号的相关性，使得网络训练获得更好的效果。模拟和实际数据的消噪实验结果表明，本文提出的消噪网络可以明显地压制微震监测资料中的相干噪声和随机噪声，极大地改善资料的信噪比。
Abstract：Surface microseismic monitoring is widely used due to easy construction and low cost, but its signal is susceptible to noise interference. In this paper, we propose a deep learning method based on Time-Frequency(T-F) domain, which constructs a U-Net training set by synchronizing the squeezing transform, and combines the migration learning strategy to eliminate the noise in the low signal-to-noise ratio data. The actual monitoring data contains a limited number of effective microseismic signals, making it difficult to construct a sufficiently large label training set. This article adopts a strategy of training using simulated datasets first, and then using actual data for transfer learning. Firstly, we perform synchrosqueezing transform on clean simulated microseismic signals and simulated microseismic signals added with real noise to obtain their high-resolution time-frequency spectra respectively, and use them to construct the time-frequency domain training set for the U-net network. After the network training is completed, a small-scale training set composed of actual microseismic monitoring signals is used to conduct transfer learning for the network and obtain the final denoising network. Using high-resolution time-frequency spectra from multiple microseismic monitoring data to construct a training set can better utilize the correlation of effective microseismic signals between monitoring data channels, resulting in better network training outcomes. The experimental results of simulation and actual data show that the denoising network proposed in this paper can significantly suppress coherent noise and random noise in microseismic monitoring data, greatly improving the signal-to-noise ratio of the data.
[1] Gómez Julián L,Velis D R.A simple method inspired by empirical mode decomposition for denoising seismic data[J].Geophysics,2016,81(6):403-413.
[2] Mousavi S M,Langston C A.Adaptive noise estimation and suppression for improving microseismic event detection[J].Journal of Applied Geophysics,2016,132:116-124.
[3] Roy V,Shukla S.Spatial and transform domain filtering method for image de-noising:A review[J].International Journal of Modern Education and Computer Science,2013,5(7):41-49.
[4] Chen Y,Ma J.Random noise attenuation by fx empirical-mode decomposition predictive filtering[J].Geophysics,2014,79(3):81-91.
[5] Huang W,Wang R,Zhang M,et al.Random noise attenuation for 3D seismic data by modified multichannel singular spectrum analysis[C].77th EAGE Conference and Exhibition 2015.Madrid,Spain:European Association of Geoscientists & Engineers,2015,2015:1-5.
[6] Herrmann F J,Hennenfent G.Non-parametric seismic data recovery with curvelet frames[J].Geophysical Journal International,2008,173(1):233-248.
[7] Liu B,Sacchi M D.Minimum weighted norm interpolation of seismic records[J].Geophysics,2004,69(6):1560-1568.
[8] 贾瑞生，赵同彬，孙红梅，等.基于经验模态分解及独立成分分析的微震信号降噪方法[J].地球物理学报，2015,58(3):1013-1023.Jia R S,Zhao T B,Sun H M,et al.Microseismic signal denoising method based on empirical mode decomposition andindependent component analysis[J].Chinese Journal of Geophysics,2015,58(3):1013-1023.
[9] Li X,Feng S,Guo Y,et al.Denoising method for microseismic signals with convolutional neural network based on transfer learning[J].International Journal of Computational Intelligence Systems,2023,16(1):91.
[10] Chai C,Maceira M,Santos-Villalobos H J,et al.Using a deep neural network and transfer learning to bridge scales for seismic phase picking[J].Geophysical Research Letters,2020,47(16):e2020GL088651.
[11] Zhang Q,Zhang W,Wu X,et al.Deep learning for efficient microseismic location using source migration-based imaging[J].Journal of Geophysical Research:Solid Earth,2022,127(3):e2021JB022649.
[12] Yao Y,Wang G,Liu L.Microseismic signal denoising using simple bandpass filtering based on normal time-frequency transform[J].Acta Geophysica,2023,71(5):2217-2232.
[13] 刘法启，张关泉.小波变换与F-K算法在滤波中的应用[J].石油地球物理勘探，1996(6):782-791.Liu F Q,Zhang G Q.Application of wavelet transform and F-K algo-rithm in filtering[J].Oil Geophysical Prospecting,1996(6):782-791.
[14] 宋维琪，刘太伟.地面微地震资料τ-p变换噪声压制[J].石油地球物理勘探，2015,50(1):48-53.Song W Q,Liu T W.Surface microseismic noise suppression with τ-p transform[J].Oil Geophysical Prospecting,2015,50(1):48-53.
[15] Huang N E,Shen Z,Long S R,et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J].Proceedings of the Royal Society of London.Series A:Mathematical,Physical and Engineering Sciences,1998,454(1971):903-995.
[16] 李稳，刘伊克，刘保金.基于稀疏分布特征的井下微地震信号识别与提取方法[J].地球物理学报，2016,59(10):3869-3882.Li W,Liu Y K,Liu B K.Downhole microseismic signal recognition and extraction based on sparse distribution leatures[J].Chinese Journal of Geophysics,2016,59(10):3869-3882.
[17] Tang N,Zhao X,Li Y,et al.Adaptive threshold shearlet transform for surface microseismic data denoising[J].Journal of Applied Geophysics,2018,153:64-74.
[18] 刘洋，Fomel Sergey,刘财，等.高阶seislet变换及其在随机噪声消除中的应用[J].地球物理学报，2009,52(8):2142-2151.Liu Y,Fomel Sergey,Liu C,et al.High-order seislet translorm and its application of random noise attenuation[J].Chinese Journal of Geophysics,2009,52(8):2142-2151.
[19] 黄忠来，张建中.同步挤压S变换[J].中国科学：信息科学，2016,46(5):643-650.Huang Z L,Zhang J Z.Synchrosqueezing S-transform[J].Scientia Sinica Informationis,2016,46(5):643-650.
[20] Ma?za Bekara,Baan M V D.Random and coherent noise attenuation by empirical mode decomposition[J].Geophysics,2009,74(5):89-98.
[21] Wu Z,Huang N E.Ensemble empirical mode decomposition:A noise-assisted data analysis method[J].Advances inAdaptive Data Analysis,2009,1(1):1-41.
[22] Torres M E,Colominas M A,Schlotthauer G,et al.A complete ensemble empirical mode decomposition with adaptive noise[C].2011 IEEE international conference on acoustics,speech and signal processing (ICASSP).Prague,Czech Republic:IEEE,2011:4144-4147.
[23] Liu W,Cao S,Chen Y.Applications of variational mode decomposition in seismic time-frequency analysis[J].Geophysics,2016,81(5):365-378.
[24] 朱权洁，隋龙琨，陈学习，等.基于EMD-SVD的矿山微震信号降噪方法及其应用[J].安全与环境工程，2024,31(3):110-119.Zhu Q J,Sui L K,Chen X X,et al.Denoising method and application of mine microseismic signal based on EMD-SVD[J].Safety and Environmental Engineering,2024,31(3):110-119.
[25] Wamriew D,Pevzner R,Maltsev E,et al.Deep neural networks for detection and location of microseismic events and velocity model inversion from microseismic data acquired by distributed acoustic sensing array[J].Sensors,2021,21(19):6627.
[26] Binder G.Neural networks for moment-tensor inversion of surface microseismic data[C].SEG International Exposition and Annual Meeting.Anaheim,USA:Society of Exploration Geophysicists,2018:SEG-2018-2998626.
[27] Chirtu M A,Radoi A.Seismic Signal Denoising using U-Net in the Time-Frequency Domain[C].2022 45th International Conference on Telecommunications and Signal Processing (TSP).Prague,Czech Republic:IEEE,2022:6-10.
[28] Ding M,Zhou Y,Chi Y.Self-attention generative adversarial network interpolating and denoising seismic signals simultaneously[J].Remote Sensing,2024,16(2):305.
[29] Daubechies I,Lu J,Wu H T.Synchrosqueezed wavelet transforms:An empirical mode decomposition-like tool[J].Applied andComputational Harmonic Analysis,2011,30(2):243-261.
[30] Mousavi S M,Langston C A,Horton S P.Automatic microseismic denoising and onset detection using the synchrosqueezed continuous wavelet transform[J].Geophysics,2016,81(4):341-355.
[31] Herrera R H,Han J,van der Baan M.Applications of the synchrosqueezing transform in seismic time-frequency analysis[J].Geophysics,2014,79(3):55-64.
[32] 秦亮，李唐律，曹脊翔，等.同步挤压S变换与τ-p变换联合的微地震信号消噪方法[J].石油地球物理勘探，2024,59(2):219-229.Qin L,Li T L,Cao J X,et al.Microseismic signal denoising method combining synchrosqueezing S-transform and τ-p transform[J].Oil Geophysical Prospecting,2024,59(2):219-229.
基本信息:
DOI：10.16441/j.cnki.hdxb.20240356
中图分类号:TE357.1;TP18
引用信息:
[1]赵欣新,林俊武,黄忠来,等.基于时-频域深度学习的水力压裂微震监测信号消噪方法[J].中国海洋大学学报(自然科学版),2025,55(12):104-114.DOI:10.16441/j.cnki.hdxb.20240356.
基金信息:
山东省自然科学基金项目(ZR2020MD047); 福建省自然科学基金项目(2022J011170)资助~~