Title: PVT properties prediction using hybrid genetic-neuro-fuzzy systems

Authors: Amar Khoukhi, Saeed Albukhitan

Addresses: Systems Engineering Department, College of Engineering and Computer Science, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia. ' Systems Engineering Department, College of Engineering and Computer Science, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia

Abstract: Pressure-volume-temperature (PVT) properties are very important in reservoir engineering computations. There are many approaches for predicting various PVT properties based on empirical correlations, statistical regression and artificial neural networks (ANNs). Unfortunately, the developed correlations are often limited and global correlations are usually less accurate compared to local correlations. In this paper, a genetic-neuro-fuzzy inference system (GANFIS) is proposed for crude oil PVT properties prediction. Simulation experiments show that the proposed technique outperforms up-to-date methods.

Keywords: pressure-volume-temperature; PVT properties; oil formation volume factor; Bob; bubble point pressure; Pb; correlation; genetic-neuro-fuzzy inference systems; GANFIS; genetic algorithms; artificial neural networks; ANNs; fuzzy logic; reservoir engineering; crude oil; simulation.

DOI: 10.1504/IJOGCT.2011.037744

International Journal of Oil, Gas and Coal Technology, 2011 Vol.4 No.1, pp.47 - 63

Published online: 29 Jan 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article