A Comparison of Self-organizing Maps and Hierarchical Cluster Analysis Approaches in Predicting Total Organic Carbon Using Intelligent Systems

Total organic carbon (TOC) is one of the main parameters for geochemical evaluation of oil and gas source rocks. In this study، we propose a two-step approach to predict total organic carbon content from well log data. Initially، the well log data are classified into a set of electrofacies (EF). The methods used to characterize and identify EF consist of self-organizing maps (SOM) and hierarchical cluster analysis (HCA). The results obtained from both methods are compared and the best method based on cluster validity tests is chosen for clustering petrophysical data into a certain number of EF. Afterwards، the TOC values are estimated from well log data by using individual artificial neural network (ANN) models constructed for each EF. In the second approach، the TOC data are estimated for the total interval by using a similar ANN model regardless of data clustering and EF determination. The results of two prediction methods are compared to each other and also with a third conventional Δlog R technique. The results show that clustering of a formation into specific units (electrofacies) provides better results in TOC prediction compared to the models constructed for the whole dataset as a single cluster. In addition، intelligent systems are more efficient than the previous conventional techniques based on Δlog R method. The proposed methodology is illustrated using a case study of the world’s largest non-associated gas reservoir، i. e. Iran South Pars Gas Field، located in the Persian Gulf.