Image feature recognition and gas permeability prediction of Gaomiaozi bentonite based on digital images and machine learning

Abstract

Gas permeability, which is measured mainly through gas permeability experiments, is a critical technical index in many engineering fields. In this study, permeability is firstly calculated based on information from a digital image and an improved permeability prediction model. The calculated results are experimentally verified. Subsequently, a self-developed image-processing program is used to extract feature parameters from a scanning electron microscopy image. Meanwhile, an extreme learning machine algorithm is used to input the image feature parameters obtained using the image-processing program into the extreme learning machine algorithm for machine learning. Additionally, we compare several typically used machine learning algorithms, which confirmed the reliability and accuracy of our algorithm. The best activation function can be obtained by comparing the predicted permeability using an appropriate number of neuron nodes. Experimental results show that the program can accurately identify the features of the microscopy image. Combining the program with an extreme learning machine neural network algorithmgas permeability results to be obtained with high accuracy. This method yields good predictions of permeability in certain cases and has been adapted to other geomaterials.

Cited as: Liu, J., Ma, S., Shen, W., Zhou, J., Hong, Y. Image feature recognition and gas permeability prediction of Gaomiaozi bentonite based on digital images and machine learning. Advances in Geo-Energy Research, 2022, 6(4): 314-323. https://doi.org/10.46690/ager.2022.04.06

References

Ahmadi, R., Shahrabi, J., Aminshahidy, B. Automatic well-testing model diagnosis and parameter estimation using artificial neural networks and design of experiments. Journal of Petroleum Exploration and Production Technology, 2017, 7: 759-783.

Alafnan, S. Utilization of supercritical carbon dioxide for mechanical degradation of organic matters contained in shales. Fuel, 2022, 316: 123427.

Arif, C., Mizoguchi, M., Setiawan, B. I., et al. Estimation of soil moisture in paddy field using artificial neural networks. International Journal of Advanced Research in Artificial Intelligence, 2012, 1: 17-21.

Bahmed, I. T., Khelifa, H., Mohamed, G., et al. Prediction of geotechnical properties of clayey soils stabilised with lime using artificial neural networks (ANNs). International Journal of Geotechnical Engineering, 2019, 13: 191-203.

Carbonell, B., Villar, M. V., Mart´ın, P. L., et al. Gas transport in compacted bentonite after 18 years under barrier conditions. Geomechanics for Energy and the Environment, 2019, 17: 66-74.

Dominguez-Olmedo, J. L., Toscano, M., Mata, J. Application of classification trees for improving optical identification of common opaque minerals. Computers & Geosciences, 2020, 140: 104480.

Dou, W., Liu, L., Jia, L., et al. Pore structure, fractal characteristics and permeability prediction of tight sandstones: A case study from yanchang formation, Ordos Basin, China. Marine and Petroleum Geology, 2021, 123: 104737.

Gebrenegus, T., Ghezzehei, T. A., Tuller, M. Physicochemical controls on initiation and evolution of desiccation cracks in sand–bentonite mixtures: X-ray CT imaging and stochastic modeling. Journal of Contaminant Hydrol-ogy, 2011, 126: 100-112.

Hossein, Y., Rasool, K., David, A., et al. Application of mathematical and machine learning models to predict differential pressure of autonomous downhole inflow control devices. Advances in Geo-Energy Research, 2021, 5(4): 386-406.

Huang, J., Zhang, Y., Sun, J., et al. Evaluation of pore size distribution and permeability reduction behavior in pervious concrete. Construction and Building Materials, 2021, 290: 123228.

Mazarei, M., Davarpanah, A., Ebadati, A., et al. The feasibility analysis of underground gas storage during an integration of improved condensate recovery processes. Journal of Petroleum Exploration and Production Technology, 2019, 9: 397-408.

Münch, B., Holzer, L. Contradicting geometrical concepts in pore size analysis attained with electron microscopy and mercury intrusion. Journal of the American Ceramic Society, 2008, 91: 4059-4067.

Shokouhi, P., Kumar, V., PrathiPati, S., et al. Physics-informed deep learning for prediction of CO2 storage site response. Journal of Contaminant Hydrology, 2021, 241: 103835.

Song, S., Liu, J., Ni, H., et al. A new automatic thresholding algorithm for unimodal gray-level distribution images by using the gray gradient information. Journal of Petroleum Science and Engineering, 2020, 190: 107074.

Song, S., Liu, J., Yang, D., et al. Pore structure characterization and permeability prediction of coal samples based on sem images-sciencedirect. Journal of Natural Gas Science and Engineering, 2019, 67: 160-171.

Taha, O. M. E., Majeed, Z. H., Ahmed, S. M. Artificial neural network prediction models for maximum dry density and optimum moisture content of stabilized soils. Transportation Infrastructure Geotechnology, 2018, 5: 146-168.

Wang, G., Qin, X., Han, D., et al. Study on seepage and deformation characteristics of coal microstructure by 3D reconstruction of CT images at high temperatures. International Journal of Mining Science and Technology, 2021, 31: 175-185.

Xiao, T., Hugh, D. Machine-learning-based object detection in images for reservoir characterization: A case study of fracture detection in shales. Leading Edge, 2018, 37: 435-442.

Xu, L., Ye, W., Chen, Y., et al. Investigation on gas permeability of compacted GMZ bentonite with consideration of variations in liquid saturation, dry density and confining pressure. Journal of Contaminant Hydrology, 2020, 230: 103622.

Zhao, J., Sun, M., Pan, Z., et al. Effects of pore connectivity and water saturation on matrix permeability of deep gas shale. Advances in Geo-Energy Research, 2022, 6(1): 54-68.