(Prediction of Condensate Gas Ratio (CGR) Using an Artificial Neural Network (ANN

Added values to project economy from condensate sales and gas deliverability loss due to condensate blockage are the main differences between gas condensate and dry gas reservoirs. To estimate the added value, one needs to obtain condensate to gas ratio (CGR); however, this needs special PVT experimental study and field tests. In the absence of experimental studies during early period of field exploration, techniques which correlate such a parameter would be of interest for engineers. Artificial Neural Network (ANN) is a multi-dimensional correlation including a large number of parameters, relating input and output data sets. Compared with an empirical correlation, an ANN model can accept more information substantially as input to the model, thereby, improving the accuracy of the predictions significantly and reducing the ambiguity of the relationship between input and output. Moreover, ANNs are fast-responding systems. Once the model has been “trained”, predictions on unknown fluids are obtained by direct and rapid calculations, without iterative computations or tuning. This paper demonstrates how ANN predicts the CGR of a gas condensate reservoir with minimum and easily accessible parameters. In development stage of the ANN model, a large number of data covering wide range of gas condensate properties and reservoir temperature were collected from the literature and National Iranian oil Company (NIOC) data bank. The qualified data set were used to train the model. The predictive ability of the model was tested using experimental data sets that were not used during the training stage. The results are in good agreement with the experimentally reported data. The proposed model exhibits sensitivity to several parameters including reservoir temperature, gas molecular weight and dew point pressure. The network has the R - square of 0.9881, 0.9837 and 0.9821 for training, validation and test, respectively.