The coefficient of isobaric thermal expansion of crude oils is essential in thermal methods of production and surface facilities design. The literature has no simple mathematical model to predict the instantaneous thermal expansion coefficient. Therefore, this study presents an artificial neural network (ANN) model and an empirical correlation for predicting crude oil's isobaric instantaneous thermal expansion coefficient. The input parameters for the ANN model and correlation are the usually measured parameters of reservoir temperature, solution gas-oil ratio, gas and oil-specific gravities, bubblepoint pressure, and pressure. The paper exclusively deals with thermal expansion for the Middle East crude oil samples. However, they should be valid for all types of crude oils with properties falling within the range of data used in this study. The statistical and graphical error analyses were used to check the performance and accuracy of the ANN model and correlation.

Keywords:
neural network, machine learning, upstream oil & gas, thermal expansion, expansion, pvt measurement, saudi arabia government, correlation, asia government, artificial intelligence
Subjects:
Fluid Characterization, Information Management and Systems, Phase behavior and PVT measurements, Neural networks
Copyright 2023, Society of Petroleum Engineers DOI 10.2118/214473-MS
You can access this article if you purchase or spend a download.