Flow regimes classification and prediction of volume fractions of the gas-oil-water three-phase flow using Adaptive Neuro-fuzzy Inference System

The used metering technique in this study is based on the dual energy (Am-241 and Cs-137) gamma ray attenuation. Two transmitted NaI detectors in the best orientation were used and four features were extracted and applied to the model. This paper highlights the application of Adaptive Neuro-fuzzy Inference System (ANFIS) for identifying flow regimes and predicting volume fractions in gas-oil-water multiphase systems. In fact, the aim of the current study is to recognize the flow regimes based on dual energy broad-beam gamma-ray attenuation technique using ANFIS. In this study, ANFIS is used to classify the flow regimes (annular, stratified, and homogenous) and predict the value of volume fractions. To start modeling, sufficient data are gathered. Here, data are generated numerically using MCNPX code. In the next step, ANFIS must be trained. According to the modeling results, the proposed ANFIS can correctly recognize all the three different flow regimes, and other ANFIS networks can determine volume fractions with MRE of less than 2% according to the recognized regime, which shows that ANFIS can predict the results precisely.