Linking fluid composition to black-oil properties

Abstract

Field development studies require, among many other things, reliable estimates of fluid properties. In cases where experimental data is unavailable or incomplete, it is necessary to predict fluid properties using correlations, based on input variables such as temperature, API, saturation pressure, gas gravity, and flash GOR. However, there are situations when even these input variables are not known, and where the only information is the initial reservoir fluid composition.The scope of the present work is to develop a predictive method which links the reservoir fluid composition to the input variables needed for the black-oil property correlations. The method is based on analysis of several experimental data sets comprising in total several hundred fluid compositions with associated black-oil properties as well as on isothermal flash calculations on 500,000 synthetic, but realistic, fluid compositions spanning a large range in compositional space. The aim of this work is therefore not to develop new correlations for oil compressibility, formation volume factors, viscosities etc. Rather, the contribution should be regarded as a novel way to incorporate reservoir fluid compositional information into an already existing framework for predicting fluid properties used for field development studies.Results from the study show that it is possible to predict API, flash gas gravity, flash GOR and saturation pressure with reasonable accuracy for a large range of reservoir fluid compositions as long as the molecular weight of the plus fraction is known. The work sheds more light on the quantitative influence of individual component mole fractions on fluid properties.