Intelligent Identification and Quantitative Characterization of Pores in Shale SEM Images Based on Pore-Net Deep-Learning Network Model
0
Citation
0
Reference
10
Related Paper
Abstract:
Among the various shale reservoir evaluation methods, the scanning electron microscope (SEM) image method is widely used. Its image can intuitively reflect the development stage of a shale reservoir and is often used for the qualitative characterization of shale pores. However, manual image processing is inefficient and cannot quantitatively characterize pores. The semantic segmentation method of deep learning greatly improves the efficiency of image analysis and can calculate the face rate of shale SEM images to achieve quantitative characterization. In this paper, the high-maturity shale of the Longmaxi Formation in the Changning area of Yibin City, Sichuan Province, and the low-maturity shale of Beibu Gulf Basin in China are studied. Based on the Pore-net network model, the intelligent identification and quantitative characterization of pores in shale SEM images are realized. The pore-net model is improved from the U-net deep-learning network model, which improves the ability of the network model to identify pores. The results show that the pore-net model performs better than the U-net model, FCN model, DeepLab V3 + model, and traditional binarization method. The problem of low accuracy of the traditional pore identification method is solved. The porosity of SEM images of high-maturity shale calculated by the pore-net network model is between 12 and 19% deviation from the experimental data. The calculated porosity of the SEM image of the low-maturity shale has a large deviation from the experimental data, which is between 14 and 47%. Compared with the porosity results calculated by other methods, the results calculated by pore-net are closer to the true value, which proves that the porosity calculated by the pore-net network model is reliable. The deep-learning semantic image segmentation method is suitable for pore recognition of shale SEM images. The fully convolutional neural network model is used to train the manually labeled shale SEM images, which can realize the intelligent recognition and quantitative characterization of the pores in the shale SEM images. It provides a certain reference value for the evaluation of shale oil and gas reservoirs and the study of other porous materials.Keywords:
Characterization
Network model
Shale Gas
Abstract In view of the weak research on the availability of typical shale oil reservoirs from the perspective of development, this study introduced a two-dimensional nuclear magnetic resonance (NMR) evaluation method on the basis of the previous one-dimensional NMR combined with centrifugal physical simulation experiments. Not only the production characteristics of typical shale oil reservoirs were studied but also the microscopic production laws of different occurrence states were studied. The results show that the pore distribution of Jilin shale is more concentrated than that of Qinghai shale. The oil of the two blocks mainly occurs in 0.01–10 ms pores, and the occurrence ratio of Jilin shale in the pores is higher, which is more than 90%. The oil production of the two blocks is mainly dominated by 0.01–10 ms pores, and the utilization efficiency contribution of these pores in Jilin shale is higher, accounting for about 80%. The utilization efficiency (UE) increases logarithmically with centrifugal force, and the growth rate of Jilin shale is greater than that of Qinghai shale. The proportion of free oil in Jilin block is less than that in Qinghai block. The shale oil in the two blocks is both at 15% final UE, and the UE of free oil in Jilin shale is about 9% and that of Qinghai shale is about 12%. The recoverability of Jilin shale is lower than that of Qinghai shale.
Shale oil extraction
Cite
Citations (1)
Measurements have been made on the chemical and physical properties of two oil shales designated as reference oil shales by the Department of Energy. One oil shale is a Green River Formation, Parachute Creek Member, Mahogany Zone Colorado oil shale from the Exxon Colony mine and the other is a Clegg Creek Member, New Albany shale from Kentucky. Material balance Fischer assays, carbon aromaticities, thermal properties, and bulk mineralogic properties have been determined for the oil shales. Kerogen concentrates were prepared from both shales. The measured properties of the reference shales are comparable to results obtained from previous studies on similar shales. The western reference shale has a low carbon aromaticity, high Fischer assay conversion to oil, and a dominant carbonate mineralogy. The eastern reference shale has a high carbon aromaticity, low Fischer assay conversion to oil, and a dominant silicate mineralogy. Chemical and physical properties, including ASTM distillations, have been determined for shale oils produced from the reference shales. The distillation data were used in conjunction with API correlations to calculate a large number of shale oil properties that are required for computer models such as ASPEN. There was poor agreement between measured and calculated molecular weights for the total shale oil produced from each shale. However, measured and calculated molecular weights agreed reasonably well for true boiling point distillate fractions in the temperature range of 204 to 399/sup 0/C (400 to 750/sup 0/F). Similarly, measured and calculated viscosities of the total shale oils were in disagreement, whereas good agreement was obtained on distillate fractions for a boiling range up to 315/sup 0/C (600/sup 0/F). Thermal and dielectric properties were determined for the shales and shale oils. The dielectric properties of the reference shales and shale oils decreased with increasing frequency of the applied frequency. 42 refs., 34 figs., 24 tabs.
Green River Formation
Oil shale gas
Shale oil extraction
Cite
Citations (10)
Cite
Citations (10)
Tight oil
Cite
Citations (6)
The concentrations and modes of occurrence of U and Th in 23 oil shale samples from China were studied.The concentrations of U and Th were determined by inductively coupled plasma-mass spectrometry.The occurrence of U and Th in Huadian and Luozigou oil shale was investigated using six-step sequential chemical extract method.The concentrations of U in oil shale are within 10×10-6 and average of 3.92×10-6.The concentrations of Th in oil shale are within 20×10-6 and average of 10.51×10-6.The abundances of U and Th in oil shale are slightly higher than in crust and close to sedimentary rocks.The experimental results of floating and sinking roughly equal that of the sequential chemical extraction in U and Th.The U and Th in oil shale are mainly existed in minerals.
Cite
Citations (0)
Retort
Oil shale gas
Shale oil extraction
Cite
Citations (3)
Cite
Citations (37)
Retort
Oil shale gas
Shale oil extraction
Synthetic crude
Cite
Citations (96)
Retorting studies of Sunbury Shale from northeastern Kentucky have been carried out systematically and compared with earlier studies of Colorado oil shale. The Fischer assay method is used as a baseline model for evaluation of the retorting parameters. The effects of inert gas flow and of particle size variation when retorting Sunbury Shale are similar to effects observed in retorting Colorado shale. However, oil yield is very dependent on shale heat-up rate; its effect is much greater on Sunbury Shale than on western shale. Pyrite has a detrimental effect on the quantity of oil generated from Sunbury Shale, and a number of design parameters impact oil production. Fischer assay is not nearly as optimum a method for obtaining oil from the Sunbury Shale as it is for retorting Colorado shale. For this eastern oil shale, an assay employing 5 times the Fischer assay heating rate and a 1380F upper temperature (i.e., 100F/min to 1380F) results in a process yield that is 110-116%
Retort
Oil shale gas
Shale oil extraction
Cite
Citations (0)