Fuzzy Logic Determination of Lithologies from Well Log Data: Application to the KTB Project Data set (Germany)

Fuzzy logic has been used for lithology prediction with remarkable success. Several techniques such as fuzzy clustering or linguistic reasoning have proven to be useful for lithofacies determination. In this paper, a fuzzy inference methodology has been implemented as a MATLAB routine and applied for the first time to well log data from the German Continental Deep Drilling Program (KTB). The training of the fuzzy inference system is based on the analysis of the multi-class Matthews correlation coefficient computed for the classification matrix. For this particular data set, we have found that the best suited membership function type is the piecewise linear interpolation of the normalized histograms; that the best combination operator for obtaining the final lithology degrees of membership is the fuzzy gamma operator; and that all the available properties are relevant in the classification process. Results show that this fuzzy logic-based method is suited for rapidly and reasonably suggesting a lithology column from well log data, neatly identifying the main units and in some cases refining the classification, which can lead to a better interpretation. We have tested the trained system with synthetic data generated from property value distributions of the training data set to find that the differences in data distributions between both wells are significant enough to misdirect the inference process. However, a cross-validation analysis has revealed that, even with differences between data distributions and missing lithologies in the training data set, this fuzzy logic inference system is able to output a coherent classification.

[1]  C.J.H. Mann,et al.  Geophysical Applications of Artificial Neural Networks and Fuzzy Logic , 2006 .

[2]  Alain Bonnet,et al.  Oil-Well Data Interpretation Using Expert System and Pattern Recognition Technique , 1983, IJCAI.

[3]  Charles L. Karr,et al.  Determination of lithology from well logs using a neural network , 1992 .

[4]  Yujin Zhang,et al.  Application of neural networks to identify lithofacies from well logs , 1999 .

[5]  Spe Member.  Comparison of an Expert System to Human Experts in Well Log Analysis and Interpretation , 1988 .

[6]  H. C. Chen,et al.  Identification of lithofacies using Kohonen self-organizing maps , 2002 .

[7]  J. Kück,et al.  The KTB Borehole—Germany’s Superdeep Telescope into the Earth’s Crust , 1995 .

[8]  János Fodor,et al.  Traditional and New Ways to Handle Uncertainty in Geology , 2001 .

[9]  Elizabeth E. Einstein,et al.  Comparison of an Expert System to Human Experts in Well-Log Analysis and Interpretation , 1990 .

[10]  Olivier Peyret,et al.  Automatic Determination of Lithology From Well Logs , 1987 .

[11]  S. K. Runcorn,et al.  Interpretation theory in applied geophysics , 1965 .

[12]  Sugio Imamura.  Integrated Interpretation of Exploration Data In Geotechnical Engineering: An Approach Using Fuzzy Theory , 1994 .

[13]  Muhammad M. Saggaf,et al.  A fuzzy logic approach for the estimation of facies from wire-line logs , 2003 .

[14]  A. B. Adeyemo,et al.  Well Log Interpretation Model for the Determination of Lithology and Fluid Contents. , 2009 .

[15]  Jan Gorodkin,et al.  Comparing two K-category assignments by a K-category correlation coefficient , 2004, Comput. Biol. Chem..

[16]  I. D. Gates,et al.  On the Capability of Support Vector Machines to Classify Lithology from Well Logs , 2010 .

[17]  A. Toumani,et al.  Automatic determination of lithology from well logs using fuzzy classification , 1994 .

[18]  Cesare Furlanello,et al.  A Comparison of MCC and CEN Error Measures in Multi-Class Prediction , 2010, PloS one.

[19]  Masoud Nikravesh,et al.  Soft Computing for Reservoir Characterization and Modeling , 2010 .

[20]  B. Matthews.  Comparison of the predicted and observed secondary structure of T4 phage lysozyme. , 1975, Biochimica et biophysica acta.

[21]  M. Rezaei,et al.  New Approach of Fuzzy Logic Applied to Lithofacies Forecasting and Permeability Values Estimation , 2009 .

[22]  Saumen Maiti,et al.  Neural network modelling and classification of lithofacies using well log data: A case study from KTB borehole site , 2007 .

[23]  S. Cuddy,et al.  Litho-Facies and Permeability Prediction from Electrical Logs using Fuzzy Logic , 2000 .

[24]  A. Bagheri,et al.  CHARACTERIZATION OF RESERVOIR ROCK TYPES IN A HETEROGENEOUS CLASTIC AND CARBONATE RESERVOIR , 2006 .

[25]  Muhammed M. Saggaf,et al.  Estimation of lithologies and depositional facies from wire-line logs , 2000 .

[26]  J. Lauterjung,et al.  The German Continental Deep Drilling Program KTB: Overview and major results , 1997 .

[27]  Michael Uschold,et al.  Building Expert Systems for Controlling Complex Programs , 1982, AAAI.

[28]  Saumen Maiti,et al.  Neural network modeling and an uncertainty analysis in Bayesian framework: A case study from the KTB borehole site , 2010 .

[29]  Swapan Chakrabarti,et al.  Comparison of four approaches to a rock facies classification problem , 2007, Comput. Geosci..

[30]  G. Wadge,et al.  Inferring the lithology of borehole rocks by applying neural network classifiers to downhole logs: an example from the Ocean Drilling Program , 1999 .

[31]  LinZsay-Shing,et al.  Lithology identification of aquifers from geophysical well logs and fuzzy logic analysis , 2005 .

[32]  Jeffrey L. Baldwin,et al.  Application Of A Neural Network To The Problem Of Mineral Identification From Well Logs , 1990 .

[33]  Randall Davis,et al.  The DIPMETER ADVISOR: Interpretation of Geologic Signals , 1981, IJCAI.

[34]  Kamy Sepehrnoori,et al.  Facies prediction from core and log data using artificial neural network technology , 1998 .

[35]  Lotfi A. Zadeh,et al.  Fuzzy Sets , 1996, Inf. Control..

[36]  Günter Zimmermann,et al.  Integrated log interpretation in the German Continental Deep Drilling Program: Lithology, porosity, and fracture zones , 1997 .

[37]  Bieng-Zih Hsieh,et al.  Lithology identification of aquifers from geophysical well logs and fuzzy logic analysis: Shui-Lin Area, Taiwan , 2005, Comput. Geosci..

[38]  Jacques Pelissier-Combescure,et al.  Faciolog - Automatic Electrofacies Determination , 1982 .

[39]  R. K. Tiwari,et al.  A Hybrid Monte Carlo Method Based Artificial Neural Networks Approach for Rock Boundaries Identification: A Case Study from the KTB Bore Hole , 2009 .

[40]  Seyed Ali Moallemi,et al.  A fuzzy logic approach for estimation of permeability and rock type from conventional well log data: an example from the Kangan reservoir in the Iran Offshore Gas Field , 2006 .

[42]  L. N. Berry,et al.  Determination of Lithology From Well Logs by Statistical Analysis , 1987 .

[43]  Paul W. J. Glover,et al.  The Application of Fuzzy Logic and Genetic Algorithms to Reservoir Characterization and Modeling , 2002 .