18 آذر 1400

محمد محمدی باغملایی

مرتبه علمی:
نشانی: -
تحصیلات: دکترای تخصصی / -
تلفن: -
دانشکده:

مشخصات پژوهش

عنوان Prediction of gas compressibility factor using intelligent models
نوع پژوهش مقالات در نشریات
کلیدواژه‌ها
ثبت نشده‌است!
مجله Natural Gas Industry B
شناسه DOI
پژوهشگران محمد محمدی باغملایی (نفر اول) ، رضا آذین (نفر دوم) ، شهریار عصفوری (نفر سوم) ، رضوان محمدی باغملایی (نفر چهارم) ، زینب زارعی (نفر پنجم)

چکیده

The gas compressibility factor, also known as Z-factor, plays the determinative role for obtaining thermodynamic properties of gas reservoir. Typically, empirical correlations have been applied to determine this important property. However, weak performance and some limitations of these correlations have persuaded the researchers to use intelligent models instead. In this work, prediction of Z-factor is aimed using different popular intelligent models in order to find the accurate one. The developed intelligent models are including Artificial Neural Network (ANN), Fuzzy Interface System (FIS) and Adaptive Neuro-Fuzzy System (ANFIS). Also optimization of equation of state (EOS) by Genetic Algorithm (GA) is done as well. The validity of developed intelligent models was tested using 1038 series of published data points in literature. It was observed that the accuracy of intelligent predicting models for Z-factor is significantly better than conventional empirical models. Also, results showed the improvement of optimized EOS predictions when coupled with GA optimization. Moreover, of the three intelligent models, ANN model outperforms other models considering all data and 263 field data points of an Iranian offshore gas condensate with R2 of 0.9999, while the R2 for best empirical correlation was about 0.8334.