ABSTRACT
Shear and Stoneley wave velocities provide useful information for petrophysical and geomechanical studies of the reservoir formation. In this study, sonic Shear and Stoneley velocities were predicted from well log data using intelligent systems including: Fuzzy logic, neural networks, neuro-fuzzy, and support vector regression. After prediction, the proposed committee machine with intelligent systems combines the first three methods in performance view. Each of these selected intelligent systems has a weight factor and the optimal combination of the weights is derived by a genetic algorithm. The study was conducted on a case study from a carbonate reservoir in South Pars gas field. The results indicate the higher performance of the committee model compared to the individual and state-of-the-art methods.
Acknowledgments
The authors extend their appreciation to the Pars Oil and Gas Company (P.O.G.C.) of Iran for data preparation and permission to publish the results of this research.
Conflict of Interest
The authors certify that they have no affiliations with or involvement in any organization with any financial or non-financial interest in the subject matter or materials discussed in this manuscript.
Additional information
Notes on contributors
Mohammad Mahdi Labani
Mohammad Mahdi Labani is a petrophysicist working in teampetro Department at DUG since finishing his PhD study in Curtin University, 2014. He has a strong academic background for petrophysical evaluation of conventional and unconventional reservoirs and did petrophysical interpretation for more than 250 wells since joining DUG.
Mostafa Sabzekar
Mostafa Sabzekar received the M.Sc. and Ph.D. degrees in computer engineering from Ferdowsi University of Mashhad, Mashhad, Iran, in 2009 and 2017, respectively. He is currently an Assistant Professor with the Department of Computer Engineering, Birjand University of Technology, Birjand, Iran. His research interests include Bioinformatics, machine learning and soft computing.