基于自编码网络的浅源和深源诱发型微震识别
|
摘要
地震波形传播的复杂多变性,导致传统互相关分析方法难以识别诱发型微震事件的深度类型.本文基于微震波形的时域、频域及时频域特征,利用自编码网络(SAE)构造具有可鉴别性的特征空间,提升对深源和浅源诱发型微震事件的分类精度.首先,针对440个诱发型微震事件,构建了大小为40的特征空间;其次,利用遗传算法(GA)和关联规则特征选择方法(CFS)对特征空间进行初步筛选,得到特征重要性程度较强的谱矩心和线性度,通过分类验证了谱矩心与震源深度有强相关性;然后,将筛选结果输入到自编码网络,采用基于无监督学习的方法获得新的特征空间;最后,利用逻辑回归(LR)对新特征空间进行交叉验证分类.与利用初步筛选的特征结果进行分类相比,利用4层的自编码网络模型对40特征进行交叉验证分类,所得正确率(Accuracy)和接收者操作特征曲线(ROC)曲线下方的面积(AUC)分别从84. 5%提高到90. 91%及84. 31%提高到87. 14%,结果表明自编码网络提高了分类模型对低能量诱发型微震事件的识别精度.
Due to the variability and complexity of propagation for seismic wave,it is difficult to identify the depth of induced microearthquakes using the traditional method based on cross-correlation.In this paper,we utilize Autoencoder Network to construct the feature space that has distinctive competence, based on time,frequency and time-frequency domain from micro-seismic waveform,which improves the accuracy of classification of micro-earthquakes.For a total of 440 events,we extract 40 features,filter them using Genetic Algorithm( GA) and Correlation-based Feature Selection.Spectral centroid and the degree of rectilinearity are important for classification. We find that there is a strong correlation between spectral centroid and depth of the source. Then, when those features are input into Autoencoder Network,we can acquire new feature space through unsupervised learning. Using Autoencoder Network with 4 layers based on cross validation,compared to use Logistic Regression( LR) in cross validation,the classification accuracy and area under the curve of ROC( AUC) based the new feature space are improved from 84. 5% to 90. 91%,84. 31% to87. 14% respectively,compared with the classification based on 40 features. Autoencoder Network improves the recognition accuracy of low energy induced micro-seismic.
Due to the variability and complexity of propagation for seismic wave,it is difficult to identify the depth of induced microearthquakes using the traditional method based on cross-correlation.In this paper,we utilize Autoencoder Network to construct the feature space that has distinctive competence, based on time,frequency and time-frequency domain from micro-seismic waveform,which improves the accuracy of classification of micro-earthquakes.For a total of 440 events,we extract 40 features,filter them using Genetic Algorithm( GA) and Correlation-based Feature Selection.Spectral centroid and the degree of rectilinearity are important for classification. We find that there is a strong correlation between spectral centroid and depth of the source. Then, when those features are input into Autoencoder Network,we can acquire new feature space through unsupervised learning. Using Autoencoder Network with 4 layers based on cross validation,compared to use Logistic Regression( LR) in cross validation,the classification accuracy and area under the curve of ROC( AUC) based the new feature space are improved from 84. 5% to 90. 91%,84. 31% to87. 14% respectively,compared with the classification based on 40 features. Autoencoder Network improves the recognition accuracy of low energy induced micro-seismic.
引文
Abu-Elsoud M A,Abou-Chadi F E Z,Amin A M,et al.2004.Classification of seismic events in suez gulf area,egypt using artificial neural network[J].International Conference on Electrical,Electronic and Computer Engineering.IEEE,46:337-340.
Allen R.V.,1978.Automatic Earthquake Recognition and Timing from Single Traces[J].Bulletin of the Seismological Society of America,68(5):1521-1532.
Bao X W,Eaton D W.2016.Fault activation by hydraulic fracturing in western Canada[J].Science,354(6318):1406-1409.
Beyreuther M,Hammer C,Wassermann J,et al.2012.Constructing a Hidden Markov Model based earthquake detector:application to induced seismicity[J].Geophysical Journal International banner,189(1):602-610.
Collobert R,Weston J.2008.A unified architecture for natural language processing:deep neural networks with multitask learning.International Conference.DBLP,:160-167.
Dan C C,Meier U,Gambardella L M,et al.2010.Deep Big Simple Neural Nets Excel on Handwritten Digit Recognition[J].Neural Computation,22(12):3207-3220.
Del Pezzo E,Esposito A,Giudicepietro F,et al.2003.Discrimination of Earthquakes and Underwater Explosions Using Neural Networks[J].Bulletin of the Seismological Society of America,93(1):215-223.
Djarfour N,Aifa T,Baddari K,et al.2008.Application of feedback connection artificial neural network to seismic data filtering[J].Comptes rendus-Geoscience,340(6):335-344.
Domingos P.1999.Meta Cost:a general method for making classifiers cost-sensitive.ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.ACM,155-164.
Dowla F U.1995.Neural Networks in Seismic Discrimination//Monitoring a Comprehensive Test Ban Treaty.Springer Netherlands,777-789.
Esposito A M,D'Auria L,Giudicepietro F,et al.2013.Neural analysis of seismic data:applications to the monitoring of Mt.Vesuvius[J].Annals of geophysics,56(4):1137-1145.
García S R,Romo M P,Mayoral J M.2006.Estimation of peak ground accelerations for Mexican subduction zone earthquakes using neural networks[J].Geofísica Internacional,46(1):51-62.
Gibbons S J,Ringdal F.2006.The detection of low magnitude seismic events using array-based waveform correlation[J].Geophysical Journal International,165(1):149-166.
Gibbons S J,Ringdal F.2006.The detection of low magnitude seismic events using array-based waveform correlation[J].Geophysical Journal International,165(1):149-166.
Gibbons S J,Srensen M B,Harris D B,et al.2007.The detection and location of low magnitude earthquakes in northern Norway using multi-channel waveform correlation at regional distances[J].Physics of the Earth&Planetary Interiors,160(3-4):285-309.
Hall M A.1999.Correlation-Based Feature Selection for Machine Learning.PhD thesis,Hamilton,New Zealand.
Hammer C,Beyreuther M,Ohrnberger M.2012.A Seismic-Event Spotting System for Volcano Fast-Response Systems[J].Bulletin of the Seismological Society of America,102(3):948-960.
Hammer C,Ohrnberger M,Fh D.2013.Classifying seismic waveforms from scratch:a case study in the alpine environment[J].Geophysical Journal International,192(1):425-439.
Hassanaitlaasri E,Akhouayri E S,Agliz D,et al.2013.Seismic Signal Classification using Multi-Layer Perceptron Neural Network[J].International Journal of Computer Applications,79(15):35-43.
Kleinbaum D.G.,Klein M.,2010.Logistic Regression:A Self LearningText,Springer[M].
Kong Q K,Allen R M,Schreier L,et al.2016.My Shake:Asmartphone seismic network for earthquake early warning and beyond[J].Science Advances,2(2):e1501055-e1501055.
Kuyuk H S,Yildirim E,Dogan E,et al.2011.An unsupervised learning algorithm:application to the discrimination of seismic events and quarry blasts in the vicinity of Istanbul[J].Natural Hazards&Earth System Sciences,11(1):93-100.
Lois A,Sokos E.Martakis,Nikos M,et al.2012.A new automatic S-onset detection technique:Application in local earthquake data[J].Geophysics.78(1):KS1-KS11.doi:10.1190/geo2012-0050.1.
Mousavi S M,Horton S P,Langston C A,et al.2016.Seismic features and automatic discrimination of deep and shallow inducedmicroearthquakes using neural network and logistic regression[J].Geophysical Journal International,207(1):ggw258.
Moya A,Irikura K.2010.Inversion of a velocity model using artificial neural networks[J].Computers&Geosciences,36(12):1474-1483.
Perry J L,Baumgardt D R.1991.Lg depth estimation and ripple fire characterization using artificial neural networks//Conference on Advances in Neural Information Processing Systems.Morgan Kaufmann Publishers Inc.544-550.
Riggelsen C,Ohrnberger M.2014.A Machine Learning Approach for Improving the Detection Capabilities at 3C Seismic Stations[J].Pure&Applied Geophysics,171(3-5):395-411.
Skidmore A K.1999.Accuracy assessment of spatial information[J]//Spatial Statistics for Remote Sensing.Springer Netherlands,197-209.
St-Onge,A.2011.Akaike information criterion applied to detecting first arrival times on microseismic data.Seg Technical Program Expanded Abstracts,4424.
Vallejos J A,Mckinnon S D.2013.Logistic regression and neural network classification of seismic records[J].International Journal of Rock Mechanics&Mining Sciences,62(9):86-95.
Vincent P,Larochelle H,Lajoie I,et al.2010.Stacked Denoising Autoencoders:Learning Useful Representations in a Deep Network with a Local Denoising Criterion[J].Journal of Machine Learning Research,11(12):3371-3408.
Wang P,Chang X,Wang Y B,et al.2014.Automatic Event Detection and Event Recovery in Low SNR Microseismic Signals Based on Time‐Frequency Sparseness[J].Chinese Journal of Geophysics,57(8):268-277,doi:10.6038/cjg20140824.
Wikipedia F.2010.Matthews correlation coefficient.Matthews Correlation Coefficient.Betascript Publishing.
Yuan J,Zhang Y J.2017.Application of Sparse Denoising Auto Encoder Network with Gradient Difference Information for Abnormal Action Detection[J].ACTA AUTOMATICA SINICA(in Chinese),43(4):604-610,doi:10.16383/j.aas.2017c150667.
Zazzaro G,Pisano F M,Romano G.2012.Bayesian Networks for Earthquake Magnitude Classification in a Early Warning System//ICDMKE 2012-International Conference on Data Mining and Knowledge Engineering.
Zhang M.2015.Earthquake Location and Detection[Ph.D.thesis].Hefei:University of Science and Technology of China.
王鹏,常旭,王一博,等.2014.基于时频稀疏性分析法的低信噪比微震事件识别与恢复[J].地球物理学报,57(08):268-277,doi:10.6038/cjg20140824.
袁静,章毓晋.2017.融合梯度差信息的稀疏去噪自编码网络在异常行为检测中的应用[J].自动化学报,43(4):604-610,doi:10.16383/j.aas.2017c150667.
张淼.2015.地震定位和检测[博士论文].合肥:中国科学技术大学.
Allen R.V.,1978.Automatic Earthquake Recognition and Timing from Single Traces[J].Bulletin of the Seismological Society of America,68(5):1521-1532.
Bao X W,Eaton D W.2016.Fault activation by hydraulic fracturing in western Canada[J].Science,354(6318):1406-1409.
Beyreuther M,Hammer C,Wassermann J,et al.2012.Constructing a Hidden Markov Model based earthquake detector:application to induced seismicity[J].Geophysical Journal International banner,189(1):602-610.
Collobert R,Weston J.2008.A unified architecture for natural language processing:deep neural networks with multitask learning.International Conference.DBLP,:160-167.
Dan C C,Meier U,Gambardella L M,et al.2010.Deep Big Simple Neural Nets Excel on Handwritten Digit Recognition[J].Neural Computation,22(12):3207-3220.
Del Pezzo E,Esposito A,Giudicepietro F,et al.2003.Discrimination of Earthquakes and Underwater Explosions Using Neural Networks[J].Bulletin of the Seismological Society of America,93(1):215-223.
Djarfour N,Aifa T,Baddari K,et al.2008.Application of feedback connection artificial neural network to seismic data filtering[J].Comptes rendus-Geoscience,340(6):335-344.
Domingos P.1999.Meta Cost:a general method for making classifiers cost-sensitive.ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.ACM,155-164.
Dowla F U.1995.Neural Networks in Seismic Discrimination//Monitoring a Comprehensive Test Ban Treaty.Springer Netherlands,777-789.
Esposito A M,D'Auria L,Giudicepietro F,et al.2013.Neural analysis of seismic data:applications to the monitoring of Mt.Vesuvius[J].Annals of geophysics,56(4):1137-1145.
García S R,Romo M P,Mayoral J M.2006.Estimation of peak ground accelerations for Mexican subduction zone earthquakes using neural networks[J].Geofísica Internacional,46(1):51-62.
Gibbons S J,Ringdal F.2006.The detection of low magnitude seismic events using array-based waveform correlation[J].Geophysical Journal International,165(1):149-166.
Gibbons S J,Ringdal F.2006.The detection of low magnitude seismic events using array-based waveform correlation[J].Geophysical Journal International,165(1):149-166.
Gibbons S J,Srensen M B,Harris D B,et al.2007.The detection and location of low magnitude earthquakes in northern Norway using multi-channel waveform correlation at regional distances[J].Physics of the Earth&Planetary Interiors,160(3-4):285-309.
Hall M A.1999.Correlation-Based Feature Selection for Machine Learning.PhD thesis,Hamilton,New Zealand.
Hammer C,Beyreuther M,Ohrnberger M.2012.A Seismic-Event Spotting System for Volcano Fast-Response Systems[J].Bulletin of the Seismological Society of America,102(3):948-960.
Hammer C,Ohrnberger M,Fh D.2013.Classifying seismic waveforms from scratch:a case study in the alpine environment[J].Geophysical Journal International,192(1):425-439.
Hassanaitlaasri E,Akhouayri E S,Agliz D,et al.2013.Seismic Signal Classification using Multi-Layer Perceptron Neural Network[J].International Journal of Computer Applications,79(15):35-43.
Kleinbaum D.G.,Klein M.,2010.Logistic Regression:A Self LearningText,Springer[M].
Kong Q K,Allen R M,Schreier L,et al.2016.My Shake:Asmartphone seismic network for earthquake early warning and beyond[J].Science Advances,2(2):e1501055-e1501055.
Kuyuk H S,Yildirim E,Dogan E,et al.2011.An unsupervised learning algorithm:application to the discrimination of seismic events and quarry blasts in the vicinity of Istanbul[J].Natural Hazards&Earth System Sciences,11(1):93-100.
Lois A,Sokos E.Martakis,Nikos M,et al.2012.A new automatic S-onset detection technique:Application in local earthquake data[J].Geophysics.78(1):KS1-KS11.doi:10.1190/geo2012-0050.1.
Mousavi S M,Horton S P,Langston C A,et al.2016.Seismic features and automatic discrimination of deep and shallow inducedmicroearthquakes using neural network and logistic regression[J].Geophysical Journal International,207(1):ggw258.
Moya A,Irikura K.2010.Inversion of a velocity model using artificial neural networks[J].Computers&Geosciences,36(12):1474-1483.
Perry J L,Baumgardt D R.1991.Lg depth estimation and ripple fire characterization using artificial neural networks//Conference on Advances in Neural Information Processing Systems.Morgan Kaufmann Publishers Inc.544-550.
Riggelsen C,Ohrnberger M.2014.A Machine Learning Approach for Improving the Detection Capabilities at 3C Seismic Stations[J].Pure&Applied Geophysics,171(3-5):395-411.
Skidmore A K.1999.Accuracy assessment of spatial information[J]//Spatial Statistics for Remote Sensing.Springer Netherlands,197-209.
St-Onge,A.2011.Akaike information criterion applied to detecting first arrival times on microseismic data.Seg Technical Program Expanded Abstracts,4424.
Vallejos J A,Mckinnon S D.2013.Logistic regression and neural network classification of seismic records[J].International Journal of Rock Mechanics&Mining Sciences,62(9):86-95.
Vincent P,Larochelle H,Lajoie I,et al.2010.Stacked Denoising Autoencoders:Learning Useful Representations in a Deep Network with a Local Denoising Criterion[J].Journal of Machine Learning Research,11(12):3371-3408.
Wang P,Chang X,Wang Y B,et al.2014.Automatic Event Detection and Event Recovery in Low SNR Microseismic Signals Based on Time‐Frequency Sparseness[J].Chinese Journal of Geophysics,57(8):268-277,doi:10.6038/cjg20140824.
Wikipedia F.2010.Matthews correlation coefficient.Matthews Correlation Coefficient.Betascript Publishing.
Yuan J,Zhang Y J.2017.Application of Sparse Denoising Auto Encoder Network with Gradient Difference Information for Abnormal Action Detection[J].ACTA AUTOMATICA SINICA(in Chinese),43(4):604-610,doi:10.16383/j.aas.2017c150667.
Zazzaro G,Pisano F M,Romano G.2012.Bayesian Networks for Earthquake Magnitude Classification in a Early Warning System//ICDMKE 2012-International Conference on Data Mining and Knowledge Engineering.
Zhang M.2015.Earthquake Location and Detection[Ph.D.thesis].Hefei:University of Science and Technology of China.
王鹏,常旭,王一博,等.2014.基于时频稀疏性分析法的低信噪比微震事件识别与恢复[J].地球物理学报,57(08):268-277,doi:10.6038/cjg20140824.
袁静,章毓晋.2017.融合梯度差信息的稀疏去噪自编码网络在异常行为检测中的应用[J].自动化学报,43(4):604-610,doi:10.16383/j.aas.2017c150667.
张淼.2015.地震定位和检测[博士论文].合肥:中国科学技术大学.