Constructing a unique two-phase compressibility factor model for lean gas condensates

Generating a reliable experimental model for two-phase compressibility factor in lean gas condensate reservoirs has always been demanding, but it was neglected due to lack of required experimental data. This study presents the main results of constructing the first two-phase compressibility factor model that is completely valid for Iranian lean gas condensate reservoirs. Based on a wide range of experimental data bank for Iranian lean gas condensate reservoirs, a unique two-phase compressibility factor model was generated using design of experiments (DOE) method and neural network technique (ANN). Using DOE, a swift cubic response surface model was generated for two-phase compressibility factor as a function of some selected fluid parameters for lean gas condensate fluids. The proposed DOE and ANN models were finally validated using four new independent data series. The results showed that there is a good agreement between experimental data and the proposed models. In the end, a detailed comparison was made between the results of proposed models.