Important Role of Adaptive Neuro – Fuzzy Inference Systems on Reservoir Facies Determination (Case Study: S-Field)
Sugiyanto, Kwartono, Neni Herawati, Anies Baasir, Yasunari Yanagiuchi
The Mantawa and Minahaki carbonate sequence is the main hydrocarbon bearing reservoir of S-Gas Field, East Sulawesi. Mantawa Formation commonly recognized as pinnacle reef build up type which were growth in the structural high areas. On the other hand, Minahaki limestone is platform carbonate type which dominantly characterized by mud supported facies. In such reservoir type, heterogeneity and reservoir distribution is the prime challenge in characterizing the reservoir.
Facies classification is part of reservoir characterization process that defines the reservoir distribution. It was carried out based on integration process between quantitative porosity-permeability relationships and qualitative core description from SCAL and routine core. This classification is usually done visually on cores and then extended to wireline logs from the cored wells. The challenge is how to apply this classification to un-cored wells based on relationships observed at the cored wells. This paper presents an approach to predict facies based on Adaptive Neuro – Fuzzy Inference Systems (ANFIS) techniques.
The ANFIS is trained on facies of cored wells based on gamma ray, density, neutron, and sonic logs. The facies selected for training the ANFIS are grouped into 4 facies representing lithological and diagenetic information, namely Packestone- Grainstone with vuggy porosity (facies 1), Packestone-Grainstone with chalky characteristic (facies 2), Wackestone-Packstone with chalky characteristic (facies 3), and mudstone with chalky characteristic (facies 4). The trained model finally used for further facies prediction on the un-cored intervals based on available logs data.
The approach of this work can be applied to fields where quantitative classification of a large number of logs by visual observation can be time-consuming and tedious. This approach can also be used to determine which logs are the most crucial for determining different types of facies. Furthermore, understanding this important process can provide an insights into future data collection.
Download full paper in here