Estimation of nuclear magnetic resonance (NMR) log permeability by integrating artificial neural network and imperialist competition optimization algorithm (ICA) in one of the oil fields of southwestern Iran

• Published in: Arabian journal of geosciences Vol. 14; no. 22
• Main Authors: Mohsenipour, Abouzar, Soleimani, Bahman, Zahmatkesh, Iman, Veisi, Iman
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.11.2021; Springer Nature B.V
• Subjects: Algorithms; Artificial neural networks; Correlation coefficient; Correlation coefficients; Earth and Environmental Science; Earth science; Earth Sciences; Evolutionary algorithms; Magnetic permeability; Mathematical analysis; Methods; Neural networks; NMR; Nuclear magnetic resonance; Oil and gas fields; Oil fields; Oil reservoirs; Optimization; Original Paper; Permeability; Predictions; Reservoirs; Scientific papers; Well logs; Neural network; Imperialist competitive algorithm; Asmari reservoir; Nuclear magnetic resonance (NMR); Permeability models
• ISSN: 1866-7511; 1866-7538
• DOI: 10.1007/s12517-021-08821-6

Permeability calculation is one of the most complex subjects in characterizing a hydrocarbon reservoir. Because among the well logs, only nuclear magnetic resonance (NMR) log is possible to do in an accurate continuous measurement way. In the present research paper, the permeability values obtained from the NMR log in the Asmari reservoir by Schlumberger or T 2 (SDR) and free fluid (Coates) methods were used to perform estimation by intelligent methods. Then, using an artificial neural network (ANN) and combining it with the imperial competitive optimal algorithm (ANN-ICA), the permeability values were predicted. Finally, predicted permeability values are compared with real values, and the prediction accuracy was regarded in view of mean squared error (MSE) and correlation coefficient ( R 2 ). The results of this study indicated that the efficiency of ANN in permeability prediction was high when this method was optimized with the imperialist competitive algorithm. This method can be used as a powerful tool in the prediction of other reservoir parameters.