Original scientific paper
Development of an artificial neural network model to predict CO2 minimum miscibility pressure
Abolfazl Bagheri Nezhad
; Department of Chemical Engineering and Petroleum, Sharif University of Technology, Tehran, Iran
S. Maryam Mousavi
; Department of Energy Engineering, Sharif University of Technology, Tehran, Iran
Sepehr Aghahoseini
; Department of Petroleum Engineering, Islamic Azad University, Kharg, Iran
Abstract
Miscible gas injection is among the most widely used enhanced oil recovery techniques, and its applications are increasingly visible in oil production worldwide. Characterizing the Minimum Miscibility Pressure (MMP) as a main parameter in these projects is a problem with no direct known solution. Available experimental methods are very time-consuming and also there is no universal method. To date, investigators have tried to find parametric correlation between different direct measurable parameters such as injected gas composition, reservoir temperature and reservoir fluid composition. However, due to complex nature of the phenomena, the proposed correlations are not accurate and reliable. Attempts are made to utilize artificial neural networks (ANNs) for identification of the relationship, which may exist between MMP, gas and reservoir fluid composition and reservoir temperature. The radial basis function (RBF) neural network architecture has been used successfully in predicting the CO2 MMP.
Keywords
minimum miscibility pressure (MMP); artificial neural network (ANN); radial basis function (RBF)
Hrčak ID:
67803
URI
Publication date:
29.4.2011.
Visits: 2.145 *