New oil fields of various regions of the Earth are characterized by significantly diverse basic properties of oil and gas, and thermobaric conditions, which in most cases are substantially high (formation temperatures reach 200°C and the reservoir pressures exceed 40-50 MPa. The difficulty of obtaining information on the main properties of oil and gas, as well as the considerable complexity of the sampling of the samples and their PVT-study, determines the need to investigate the correlation relationships between the individual properties of oil and gas. The objective of this study is to obtain correlation based on neural networks for Iranian oil fields, which differ not only in terms of their thermobaric conditions, but also in the basic properties of their reservoir fluids. A new mathematical model is proposed using machine learning techniques for estimating PVT fluids properties such as bubble pressure and oil formation volume factor as a function of the solution gas-oil ratio, gas density, oil density, and temperature. The result obtained with this new approach is compared with previous published correlations. The model was based on artificial neural networks, and developed using 180 published data sets from the Iran. This improvement in PVT calculation accuracy will be of invaluable support for simulations and designs applied in Oil industry.
1. Aleksander I., Morton H., An introduction to neural computing, London: London: Chapman & Hall, 1990, 218 p.
2. Al-Marhoun M.A., New correlations for formation volume factors of oil and gas mixtures, J. Can. Pet. Technol., 1992, V. 31, pp. 22–26.
3. Al-Marhoun M.A., Evaluation of empirically derived PVT properties for Middle East crude oils, J. Pet. Sci. Eng., 2003, V. 42, pp. 209–221.
4. Standing M.B., A pressure-volume-temperature correlation for mixtures of California oils and gasses, In: Drilling and Production Practice, Tulsa: American Petroleum Institute, 1947, pp. 275–278.