A Novel Empirical Correlation to Predict the Dew Point Pressure using Intelligent Algorithms

Abstract

Dew point is an important thermodynamic parameter for a gas condensate reservoir and has a very complicated nature due to its reliance on the composition of the mixture. For accurate prediction of this property, it is imperative to develop accurate models that are not computationally expensive. Currently, there exists various methodologies to estimate dew point pressure at various temperatures and hydrocarbon compositions. These methods include equation of states (EOS), analytical methods, and empirical correlations. These methods, however, have limitations in terms of accuracy or computational expense. This paper proposes a new empirical correlation to predict the dew point pressure for gas condensate reservoirs utilizing computational intelligence algorithms, namely Artificial Neural Networks (ANN), Functional Networks (FN), and Support Vector Machines (SVM). The available data set comprises of dew-point pressure, temperature, component mole fractions (C7+, CH4, N2, CO2, and H2S). This data is divided into two parts to allow for the training and development of the new model and testing/validation phase. For ANN model the weights and bias as well as the neurons in the hidden layer are tuned to result in an optimized model. In the FN model, a number of learning algorithms were tested to reach to the optimum model to get accurate results. For SVM, three main parameters were explored to develop the intelligent model which includes epsilon, kernel parameters and ‘C’. This study has resulted in the development of an empirical equation that is able to predict the dew point pressure accurately consuming the least amount of computation time. The proposed equation can be applied for a variable range of composition and temperature/ pressure conditions. This is done by incorporating the effect of composition through two equations for normal-boiling point condition as well as the critical-temperature of the mixture. Model accuracy has been validated through a comparative analysis incorporating actual experimental data from various gas-condensate reservoir samples. These data-set includes various published sources and the results of Wilson, Whitson, and EOS. This work showcases the effectiveness of intelligent models in providing answers with the least amount of error. A comparative analysis for the various computational models is done to come up with a correlation proving accurate results for the dew point pressure. The proposed correlation can predict the output with relatively small errors.

A Machine Learning Based Accelerated Approach to Infer the Breakdown Pressure of the Tight Rocks

“…Machine learning and artificial intelligence is making great progress in the oil and gas industry, with various researchers investigating advance applications pertaining to complex/heterogenous systems (Konoshonkin et al, 2020;Ma et al, 2018;Mohaghegh et al, 1994;wu zy and liu, 2018;Zhang et al, 2005Zhang et al, , 2012 (Elkatatny et al, 2016;Tariq, 2018;Tariq et al, 2021Tariq et al, , 2020cTariq et al, , 2020dTariq et al, , 2019aTariq et al, , 2019bTariq and Mahmoud, 2019). A wide range of ML algorithms have been employed to develop models/correlations for estimating various parameters related to hydrocarbons development (Ahmadi et al, 2014;Anifowose et al, 2015;da Silva et al, 2005;Janjua et al, 2016;Khan et al, 2019aKhan et al, , 2019bKhan et al, , 2018bKhan et al, , 2018aKhan et al, , 2018cLi et al, 2020;Tariq et al, 2018Tariq et al, , 2016Tohidi-Hosseini et al, 2016). Lithology determination from well logs utilizing neural networks has been the subject for various publications (Amir et al, 2020;Rogers et al, 1992).…”

An Effective Method of Estimating Nuclear Magnetic Resonance Based Porosity Using Deep Learning Approach

“…Machine learning is making great progress in the oil and gas industry, with various researchers investigating advance applications pertaining to complex/heterogenous systems (Konoshonkin et al, 2020;Ma et al, 2018;Mohaghegh et al, 1994;wu zy and liu, 2018;Zhang et al, 2005Zhang et al, , 2012 (Elkatatny et al, 2016;Tariq, 2018;Tariq et al, 2021Tariq et al, , 2019aTariq et al, , 2019bZeeshan Tariq et al, 2020b, 2020cTariq and Mahmoud, 2019). A wide range of AI algorithms have been employed to develop models/correlations for estimating various parameters related to hydrocarbons development (Ahmadi et al, 2014;Anifowose et al, 2015;da Silva et al, 2005;Janjua et al, 2016;Khan et al, 2019aKhan et al, , 2019bKhan et al, , 2018bKhan et al, , 2018aKhan et al, , 2018cLi et al, 2020;Tariq et al, 2018Tariq et al, , 2016Tohidi-Hosseini et al, 2016). Lithology determination from well logs utilizing neural networks has been the subject for various publications (Amir et al, 2020;Rogers et al, 1992).…”