Seismic facies analysis based on self-organizing map and empirical mode decomposition

Abstract Seismic facies analysis plays an important role in seismic interpretation and reservoir model building by offering an effective way to identify the changes in geofacies inter wells. The selections of input seismic attributes and their time window have an obvious effect on the validity of classification and require iterative experimentation and prior knowledge. In general, it is sensitive to noise when waveform serves as the input data to cluster analysis, especially with a narrow window. To conquer this limitation, the Empirical Mode Decomposition (EMD) method is introduced into waveform classification based on SOM. We first de-noise the seismic data using EMD and then cluster the data using 1D grid SOM. The main advantages of this method are resolution enhancement and noise reduction. 3D seismic data from the western Sichuan basin, China, are collected for validation. The application results show that seismic facies analysis can be improved and better help the interpretation. The powerful tolerance for noise makes the proposed method to be a better seismic facies analysis tool than classical 1D grid SOM method, especially for waveform cluster with a narrow window.

[1]  Gabriel Rilling,et al.  Empirical mode decomposition as a filter bank , 2004, IEEE Signal Processing Letters.

[2]  N. Huang,et al.  The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , 1998, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[3]  Mrinal K. Sen,et al.  Artificial immune based self organizing maps for seismic facies analysis , 2012 .

[4]  Ali Moradzadeh,et al.  Classification and identification of hydrocarbon reservoir lithofacies and their heterogeneity using seismic attributes, logs data and artificial neural networks , 2012 .

[6]  Ren-fei Tian,et al.  EMD and Teager–Kaiser energy applied to hydrocarbon detection in a carbonate reservoir , 2014 .

[7]  Kurt J. Marfurt,et al.  Characterizing a Mississippian Tripolitic Chert reservoir using 3D unsupervised seismic facies analysis and well logs: an example from Osage County, Oklahoma , 2012 .

[8]  Larry W. Lake,et al.  Seismic Facies Identification and Classification Using Simple Statistics , 2008 .

[9]  M. Matos,et al.  Unsupervised seismic facies analysis using wavelet transform and self-organizing maps , 2007 .

[10]  Ya-juan Xue,et al.  A comparative study on hydrocarbon detection using three EMD-based time–frequency analysis methods , 2013 .

[11]  M. Ehrhardt,et al.  Evaluation of decomposition tools for sea floor pressure data: A practical comparison of modern and classical approaches , 2012, Comput. Geosci..

[12]  Bradley Matthew Battista,et al.  Application of the Empirical Mode Decomposition and Hilbert-Huang Transform to Seismic Reflection Data , 2007 .

[13]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[14]  M. Nafi Toksöz,et al.  Seismic facies classification and identification by competitive neural networks , 2003 .

[15]  Kurt J. Marfurt,et al.  Applying Self-organizing Maps of Multiple Attributes, an Example From the Red-Fork Formation, Anadarko Basin , 2010 .

[16]  Reservoir detection based on EMD and correlation dimension , 2009 .

[17]  John P. Castagna,et al.  Support Vector Machine (SVM) pattern recognition to AVO classification , 2004 .

[18]  M. T. Taner,et al.  Reservoir characterization by calibration of self‐organized map clusters , 2001 .

[19]  Long Jin,et al.  Fusion Based Classification Method And Its Application , 2007 .

[20]  Thierry Coléou,et al.  Interpreter's Corner—Unsupervised seismic facies classification: A review and comparison of techniques and implementation , 2003 .

[21]  Michelle Chaves Kuroda,et al.  Interpretation of seismic multiattributes using a neural network , 2012 .