Technical Screening of Enhanced Oil Recovery Methods - A Case Study of Block C in Offshore Angolan Oilfields

Abstract

This paper presents a technical screening of enhanced oil recovery (EOR) methods by using an Artificial Intelligence (AI) model based on neuro-fuzzy (NF) algorithm. The presented NF approach will enable the user to select a suitable EOR method based on available worldwide successful EOR data and field under investigation. The NF approach presented in this study is a five layered feedforward-backpropagation neural networks where the knowledge pattern is extracted by combining both the searching potential of fuzzy-logic and the learning capability of neural network to make a priori decision. The extracted knowledge from the NF system can be expressed in the form of fuzzy rules by computing weights, number of rules and fuzzy set parameters and validated against reservoir properties data trained from worldwide successful EOR projects. The successfully trained and validated model is then tested on the Angolan oilfield data (Block C) where EOR application is yet to be fully established. The test results show that the NF presented in this study can be used for technical selection of suitable EOR techniques. Within the area investigated (Block C) polymer, hydrocarbon gas, and combustion were identified as the suitable techniques.