Search

The coal reservoirs have low mechanical strength and strong stress sensitivity. The study of stress sensitivity is of great significance to the efficient development of coalbed methane (CBM). In order to reveal the stress and strain characteristics of coal reservoirs, the low-field nuclear magnetic resonance (NMR) experiments with variable confining pressures were carried out on three coal samples from China. Based on the T2 spectrum of coal samples under different confining pressures, the pore distribution, connectivity, and porosity variation rules of the coal samples were analyzed. Results show that the peaks of mesopore, macropore-fissure in the coal samples are well developed; however, the micropore peaks are poorly developed. There are differences in the stress deformation of pores and fractures in coals. With increasing confining pressures from 0 to 16 MPa, the peak of fractures decrease most obvious, and the peak of mesopore and macropore-fissure are relatively obvious, while the peak of the small pores is not significantly reduced. With the confining pressure increases, the total NMR porosity of coal decreases as the rule of negative exponential function, which is mainly controlled by changes of the porosity of seepage pore. The effect of adsorption pore on porosity changes is not obvious. [ABSTRACT FROM AUTHOR]

Based on the basic principle of the porosity method in image segmentation, considering the relationship between the porosity of the rocks and the fractal characteristics of the pore structures, a new improved image segmentation method was proposed, which uses the calculated porosity of the core images as a constraint to obtain the best threshold. The results of comparative analysis show that the porosity method can best segment images theoretically, but the actual segmentation effect is deviated from the real situation. Due to the existence of heterogeneity and isolated pores of cores, the porosity method that takes the experimental porosity of the whole core as the criterion cannot achieve the desired segmentation effect. On the contrary, the new improved method overcomes the shortcomings of the porosity method, and makes a more reasonable binary segmentation for the core grayscale images, which segments images based on the actual porosity of each image by calculated. Moreover, the image segmentation method based on the calculated porosity rather than the measured porosity also greatly saves manpower and material resources, especially for tight rocks. [ABSTRACT FROM AUTHOR]