Prediction of permeability of tight sandstones from mercury injection capillary pressure tests assisted by a machine-learning approach
Abstract view|0|times PDF download|0|times
Abstract
Mercury injection capillary pressure analysis is a methodology for determining different petrophysical properties, including bulk density, porosity, and pore throat distribution. In this work, distinct parameters derived from mercury injection capillary pressure tests was considered for the prediction of permeability by coupling machine learning and theoretical approaches in a dataset composed of 246 tight sandstone samples. After quality checking the dataset, the feature selection was carried out by correlation analysis of different theoretical permeability models and statistical parameters with the measured permeability. Finally, porosity, median capillary pressure, Winland model, and mean pore-throat radius (corresponding to the saturation range 0.4-0.8) were chosen as the input features of the machine learning model. As the machine learning approach, a support vector machine (SVM) model with a radial basis function kernel was proposed. Furthermore, the model and its metaparameters were trained with a particle swarm optimization (PSO) algorithm to avoid over-fitting or under-fitting. In contradiction to the theoretical models, the implemented SVM-PSO model could acceptably predict the experimentally measured permeability values with an R2 rate of over 0.88 for training and testing datasets. The introduced approach could reduce the mean relative errors from about 10 to values less than 0.45. The improvements were more significant for low permeability samples. This successful implementation shows the potential of coupled usage of theoretical and machine learning methodologies for improved prediction of permeability of tight sandstone rocks.
Cited as: Abbasi, J., Zhao, J., Ahmed, S., Jiao, L., Andersen, P., Cai, J. Prediction of permeability of tight sandstones from mercury injection capillary pressure tests assisted by a machine-learning approach. Capillarity, 2022, 5(5): 91-104. https://doi.org/10.46690/capi.2022.05.02
Keywords
Full Text:
PDFReferences
Abbasi, J., Andersen, P. Ø. Theoretical comparison of two setups for capillary pressure measurement by centrifuge. Heliyon, 2022, 8(9): e10656.
Ahmad, N., W¨orman, A., Sanchez-Vila, X., et al. Injection of CO2-saturated brine in geological reservoir, A way to enhanced storage safety. International Journal of Greenhouse Gas Control, 2016, 54: 129-144.
Arabjamaloei, R., Daniels, D., Ebeltoft, E., et al. Validation of permeability and relative permeability data using mercury injection capillary pressure data. E3S Web of Conferences, 2019, 89: 01001.
Avriel, M. Nonlinear Programming: Analysis and Methods. Massachusetts, USA, Courier Corporation, 2003.
Blunt, M. J. Multiphase Flow in Permeable Media. London, UK, Cambridge University Press, 2017.
Burn, R. P., Mandelbrot B. B. The Fractal Geometry of Nature. New York, USA, W. H. Freeman, 1984.
Changtao, Y., Shuyuan, L., Hailong, W., et al. Pore structure characteristics and methane adsorption and desorption properties of marine shale in Sichuan Province, China. RSC Advances, 2018, 8: 6436-6443.
Choi, S., Kim, T., Yu, W. Performance evaluation of RANSAC family. Paper Presented at British Machine Vision Conference, London, UK, 7-10 September, 2009.
Cortes, C., Vapnik, V. Support-vector networks. Machine Learning, 1995, 20: 273-297.
Croux, C., Dehon, C. Influence functions of the spearman and kendall correlation measures. Statistical Methods & Applications, 2010, 19: 497-515.
Erofeev, A., Orlov, D., Ryzhov, A., et al. Prediction of porosity and permeability alteration based on machine learning algorithms. Transport in Porous Media, 2019, 128: 677-700.
Eslami, M., Kadkhodaie-Ilkhchi, A., Sharghi, Y., et al. Construction of synthetic capillary pressure curves from the joint use of NMR log data and conventional well logs. Journal of Petroleum Science and Engineering, 2013, 111: 50-58.
Fan, C., Zhong, C., Zhang, Y., et al. Geological factors controlling the accumulation and high yield of marine-Facies shale gas: Case study of the wufeng-longmaxi formation in the dingshan area of southeast sichuan, China. Acta Geologica Sinica, 2019, 93: 536-560.
Feng, F., Wang, P., Wei, Z., et al. A new method for predicting the permeability of sandstone in deep reservoirs. Geofluids, 2020, 2020: 8844464.
Ge, X., Fan, Y., Deng, S., et al. An improvement of the fractal theory and its application in pore structure evaluation and permeability estimation. Journal of Geophysical Research: Solid Earth, 2016, 121(9): 6333-6345.
Guo, B., Ghalambor, A., Duan, S. Correlation between sandstone permeability and capillary pressure curves. Journal of Petroleum Science and Engineering, 2004, 43(3-4): 239-246.
Hébert, V., Porcher, T., Planes, V., et al. Digital core repository coupled with machine learning as a tool to classify and assess petrophysical rock properties. E3S Web of Conferences, 2020, 146: 01003.
Jiao, L., Andersen, P. Ø., Zhou, J., et al. Applications of mercury intrusion capillary pressure for pore structures: A review. Capillarity, 2020, 3(4): 62-74.
Jones, S. C. A technique for faster pulse-decay permeability measurements in tight rocks. SPE Formation Evaluation, 1997, 12(1): 19-25.
Kameyama, K. Particle swarm optimization-a survey. IEICE Transactions on Information and Systems, 2009, 92(7): 1354-1361.
Karniadakis, G. E., Kevrekidis, I. G., Lu, L., et al. Physics-informed machine learning. Nature Reviews Physics, 2021, 3(6): 422-440.
Karpatne, A., Ebert-Uphoff, I., Ravela, S., et al. Machine Learning for the Geosciences: Challenges and opportunities. IEEE Transactions on Knowledge and Data Engineering, 2019, 31(8): 1544-1554.
Kendall, M. G. Rank Correlation Methods 4th edn. London, UK, High Wycombe, 1976.
Kennedy, J., Eberhart, R. Particle swarm optimization. Paper Presented at Proceedings of ICNN’95-International Conference on Neural Networks, Perth, WA, 27 November-1 December, 1995.
Kolodzie, S. Analysis of pore throat size and use of the waxman-Smits equation to determine ooip in spindle field, colorado. Paper SPE-9382-MS Presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, 21-24 September, 1980.
Li, K. Analytical derivation of brooks-corey type capillary pressure models using fractal geometry and evaluation of rock heterogeneity. Journal of Petroleum Science and Engineering, 2010, 73(1-2): 20-26.
Lin, Q., Bijeljic, B., Pini, R., et al. Imaging and measurement of pore-scale interfacial curvature to determine capillary pressure simultaneously with relative permeability. Water Resources Research, 2018, 54(9): 7046-7060.
Liu, M., Xie, R., Li, C., et al. Determining the segmentation point for calculating the fractal dimension from mercury injection capillary pressure curves in tight sandstone. Journal of Geophysics and Engineering, 2018, 15(4): 1350-1362.
Liu, Y., Xian, C., Li, Z., et al. A new classification system of lithic-rich tight sandstone and its application to diagnosis high-quality reservoirs. Advances in Geo-Energy Research, 2020, 4(3): 286-295.
McPhee, C., Reed, J., Zubizarreta, I. Core Analysis: A Best Practice Guide. Amsterdam, Netherlands, Elsevier, 2015.
Menke, H. P., Maes, J., Geiger, S. Upscaling the porosity-permeability relationship of a microporous carbonate for darcy-scale flow with machine learning. Scientific Reports, 2021, 11(1): 2625.
Purcell, W. R. Capillary pressures-their measurement using mercury and the calculation of permeability therefrom. Journal of Petroleum Technology, 1949, 1: 39-48.
Puth, M. T., Neuh¨auser, M., Ruxton, G. D. Effective use of spearman’s and kendall’s correlation coefficients for association between two measured traits. Animal Behaviour, 2015, 102: 77-84.
Rezaee, R., Saeedi, A., Clennell, B. Tight gas sands permeability estimation from mercury injection capillary pressure and nuclear magnetic resonance data. Journal of Petroleum Science and Engineering, 2012, 88: 92-99.
Smola, A. J., Schölkopf, B. A tutorial on support vector regression. Statistical Computations, 2004, 14(3): 199-222.
Swanson, B. F. Simple correlation between permeabilities and mercury capillary pressures. Journal of Petroleum Technology, 1981, 33(12): 2498-2504.
Tran, H., Sakhaee-Pour, A., Bryant, S. L. A simple relation for estimating shale permeability. Transport in Porous Media, 2018, 124(3): 883-901.
Wang, F., Jiao, L., Liu, Z., et al. Fractal analysis of pore structures in low permeability sandstones using mercury intrusion porosimetry. Journal of Porous Media, 2018, 21(11): 1097-1119.
Wang, F., Yang, K., You, J., et al. Analysis of pore size distribution and fractal dimension in tight sandstone with mercury intrusion porosimetry. Results in Physics, 2019, 13: 102283.
Wang, J., Wang, J. Low-amplitude structures and oil-gas enrichment on the yishaan slope, ordos basin. Petroleum Exploration and Development, 2013, 40(1): 52-60.
Xiao, L., Liu, D., Wang, H., et al. The applicability analysis of models for permeability prediction using mercury injection capillary pressure (MICP) data. Journal of Petroleum Science and Engineering, 2017, 156: 589-593.
Xiao, L., Liu, X., Zou, C., et al. Comparative study of models for predicting permeability from nuclear magnetic resonance (NMR) logs in two Chinese tight sandstone reservoirs. Acta Geophysica, 2014, 62(1): 116-141.
Zhang, C., Cheng, Y., Zhang, C. An improved method for predicting permeability by combining electrical measurements and mercury injection capillary pressure data. Journal of Geophysics and Engineering, 2017, 14(1): 132-142.
Zhang, G., Wang, Z., Mohaghegh, S., et al. Pattern visualization and understanding of machine learning models for permeability prediction in tight sandstone reservoirs. Journal of Petroleum Science and Engineering, 2021, 200: 108142.
Zhao, G., Li, X., Liu, M., et al. Reservoir characteristics of tight sandstone and sweet spot prediction of dibei gas field in eastern kuqa depression, northwest China. Energies, 2022, 15(9): 3135.
Refbacks
- There are currently no refbacks.
Copyright (c) 2022 The Author(s)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.