On the Capability of Support Vector Machines to Classify Lithology from Well Logs

• Published in: Natural resources research (New York, N.Y.) Vol. 19; no. 2; pp. 125 - 139
• Main Authors: Al-Anazi, A, Gates, I. D
• Format: Journal Article
• Language: English
• Published: Boston Boston : Springer US 01.06.2010; Springer US; Springer; Springer Nature B.V
• Subjects: Chemistry and Earth Sciences; Classification; Classifiers; Computer Science; core data; Data points; Discriminant analysis; Earth and Environmental Science; Earth Sciences; Earth, ocean, space; Empirical analysis; Exact sciences and technology; facies classification; Fossil Fuels (incl. Carbon Capture); Geography; Learning theory; Lithology; log data; Logs; Mathematical Modeling and Industrial Mathematics; Mathematical models; Metallic and non-metallic deposits; Mineral Resources; Neural networks; Optimization; Physics; probabilistic neural networks; Risk; Risk reduction; Sandstone; Statistical analysis; Statistical methods; Statistics; Statistics for Engineering; Support vector machines; Sustainable Development; Training; Well logs; Support vector machines; probabilistic neural networks; core data; log data; discriminant analysis; facies classification; models; testing; lithology; neural networks; reservoirs; well logs; classification; sandstone; sedimentary rocks; drill cores; lithofacies; performances; frame structure; errors; theory; clastic rocks
• ISSN: 1520-7439; 1573-8981
• DOI: 10.1007/s11053-010-9118-9

The increasing technical demands placed on models with promising generalization performance to identify lithology from well-log data points has led to search for more efficient methods than conventional statistical methods. Conventional methods such as discriminant analysis recently applied neural networks belong to the general class of Empirical Risk Minimization techniques which aim to minimize training error. On the other hand, methods built on support vector machines (SVMs) are based on the Structural Risk Minimization principle which in turn is based on statistical learning theory. Statistical learning enhancement gives better generalization abilities by minimizing the testing error. In this research, a new modeling framework based on SVMs is described to identify lithfacies from well logs. A SVM classification formulation is presented together with feature selection based on fuzzy theory to identify potential features to enable discriminatory power from well logs. The results demonstrate that SVMs provide an accurate means to identify lithofacies in a heterogeneous sandstone reservoir. The true classification was based on detailed core description of a training well. The SVM-based lithology classifier was compared to discriminant analysis and probabilistic neural networks. The results reveal that the SVM classifier performs the best of the three methods.