Electrical rock typing using Gaussian mixture model to determine cementation factor

Abstract Many studies have worked on the estimation of fluid saturation as an important petrophysical property in hydrocarbon reservoirs. Based Archie's law, proper determination cementation factor ( m ) can lead to accurate values water saturation. Given that is mainly affected by electrical properties rock, quality index (EQI) be used estimate through a novel rock typing technique. Despite efficient applicability EQI for classification rocks, with similar behaviors, into distinct types (ERTs), manual implementation this method time-consuming and gets excessively more difficult larger datasets. In work, fast automated version methodology was presented. As fuzzy clustering algorithm, Gaussian mixture model (GMM) implemented large quantity carbonate sandstone samples cluster them ERTs based values. To end, 100 data points were randomly selected testing purposes, remaining training subsets samples. An innovative hybrid EQI-GMM approach developed determine optimum number clusters. Furthermore, results two commonly-used criteria, namely Schwarz's Bayesian Criterion (BIC) Akaike Information (AIC), showed they fail specify properly. The predicted (RMSE 0.0167 0.0056 samples, respectively) than outputs traditional Archie’s law 1.6697 0.1850 respectively).