Version 1
: Received: 27 June 2019 / Approved: 9 July 2019 / Online: 9 July 2019 (04:53:33 CEST)
How to cite:
Shamshirband, S.; Baghban, A.; Sasanipour, J.; Hadipoor, M. On the Investigation of Temperature Effects on Oil Relative Permeability: Robust Modeling and Data Assessments. Preprints2019, 2019070129. https://doi.org/10.20944/preprints201907.0129.v1.
Shamshirband, S.; Baghban, A.; Sasanipour, J.; Hadipoor, M. On the Investigation of Temperature Effects on Oil Relative Permeability: Robust Modeling and Data Assessments. Preprints 2019, 2019070129. https://doi.org/10.20944/preprints201907.0129.v1.
Cite as:
Shamshirband, S.; Baghban, A.; Sasanipour, J.; Hadipoor, M. On the Investigation of Temperature Effects on Oil Relative Permeability: Robust Modeling and Data Assessments. Preprints2019, 2019070129. https://doi.org/10.20944/preprints201907.0129.v1.
Shamshirband, S.; Baghban, A.; Sasanipour, J.; Hadipoor, M. On the Investigation of Temperature Effects on Oil Relative Permeability: Robust Modeling and Data Assessments. Preprints 2019, 2019070129. https://doi.org/10.20944/preprints201907.0129.v1.
Abstract
Various empirical models are available to evaluate the temperature effects on relative permeability of the different rock and fluid systems. However, the implementation of limited experimental data points may hinder the applicability of such models to other systems. This study aims to develop new predictive models for kroestimation based on multilayer perceptron artificial neural network (MLP-ANN), adaptive neuro-fuzzy inference system (ANFIS), and least squares support vector machine (LSSVM) approaches. A database comprising of 626 data points applied to the model development. The independent variables are temperature, oil viscosity, water viscosity, water saturation ( ), and the absolute permeability. Each variable covers a wide range of variations which increases models’ potential to be applied in various systems with different characteristics. The doubtful experimental data points excluded using a leverage value approach and a sensitivity analysis carried out to determine the quantitative impact of every individual independent variable on the kro. Statistical error analyses demonstrated the coefficient of determination (R2) values of 0.985, 0.975, and 0.999 for MLP-ANN, ANFIS, and LSSVM, respectively. The comparative study indicated that the LSSVM had the best performance regarding both graphical and statistical error analyses among the newly proposed models and previously reported models in the literature.
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
