A neural computing strategy to estimate dew-point pressure of gas condensate reservoirs
Reza Daneshfar, Farhad Keivanimehr, Mohammad Mohammadi‐Khanaposhtani, Alireza Baghban
Abstract
A novel multilayer perceptron artificial neural network (MLP-ANN) model is proposed to estimate the dew-point pressure (DPP) of gas condensate reservoirs as a function of gas composition, reservoir temperature and, molecular weight and specific gravity of C7+. For this purpose, a comprehensive database was prepared by reviewing literature and the results of MLP-ANN are graphically and statistically compared with these actual values. The R-squared (R2) and mean relative error are determined to be 0.9868 and 1.5%, respectively, which reveals that DPP values are well predicted by this model. Furthermore, the MLP-ANN model is compared with previous developed models.