Full text loading...
-
Estimation of Pore Types in a Carbonate Reservoir through Artificial Neural Networks
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 81st EAGE Conference and Exhibition 2019, Jun 2019, Volume 2019, p.1 - 5
Abstract
During the past years, machine learning techniques such as artificial neural networks (ANNs) have been widely used in many disciplines for solving a wide range of complex problems. However, application of such techniques in reservoir characterization for the purpose of, for example, pore type estimation has not been understood. Rock physical modeling of carbonate rocks has encountered numerous problems due to the complexity of the pore system in such rocks. In this regard, ANNs could be helpful when huge amounts of data are available. Therefore, in this study, ANN was used to estimate pore type in a carbonate reservoir in southwest of Iran, as a case study. For this purpose, a total of 371 samples were gathered from petrophysical logs of three wells penetrated into a given carbonate formation. Then, following an experimental approach, these samples were classified into four different categories based on the observed pore types. The model was trained on 240 data points and the remaining portion of data set was reserved for validation purposes. Validation of the estimated results against the corresponding experimental data indicated that ANNs can be seen as a practical tool for pore type estimation in carbonate rocks with high accuracy.