The stress sensitivity of reservoirs plays an important role in the field of oil and gas exploration and development. As an effective stress sensitivity analysis method, digital core has the problem of excessive computational cost. We thus propose a low computational cost method for predicting core stress sensitivity. This paper first established digital cores of carbonate rocks under different effective stresses, then calculated permeability under different stresses by a new decoupling method. It simplified the three-dimensional pore-scale simulation to multiple decoupled two-dimensional ones and greatly reduced the computational complexity. Meanwhile, to solve the problem of high computational cost, we upscaled the CT images and tested the effect of image resolutions on the accuracy of the stress sensitivity analysis. The results show that the stress sensitivity of digital core image is similar to that of the original image after upscaling. We thus can use scaled image for the calculation of the stress sensitivity of core. This computationally effective method provides a basis for stress sensitivity prediction of tight lithology reservoirs.
Reservoir stress sensitivity is an important research topic in the field of oil and gas exploration and exploitation. During oil and gas development, the pressure on the rock skeleton increases as the fluid in the reservoir space where the oil and gas migrate decreases. Pores, throats, and microfractures in the reservoir are compressed or destroyed, resulting in reduced permeability. This phenomenon is called the stress sensitivity of reservoir (Cao and Lei, 2019; Hu et al., 2020; Yang et al., 2021). In recent years, with the increasing proportion of unconventional oil and gas production, the study of reservoir stress sensitivity plays an increasingly important role in seismic interpretation, well-logging reservoir evaluation, reservoir stimulation, and oil and gas development (Wang et al., 2017).
As a well-established method, pressure sensitive macroscopic experiments have been widely used. The loading method of stress sensitive macroscopic experiment is mainly divided into two categories: constant internal pressure with variable confining pressure and constant confining pressure with variable internal pressure (pore pressure). The loading method of constant confining pressure with variable internal pressure is to keep the confining pressure constant during the experiment, while gradually reducing the internal pressure. Then measure the permeability corresponding to different pressures, and obtain the stress sensitivity of permeability (Reyes and Osisanya, 2002). In comparison, the loading method of constant internal pressure with variable confining pressure is the opposite. During the development of oil and gas reservoirs, the pressure in the overlying formations is usually constant. The pore pressure gradually decreases with development, which alters the effective pressure. Therefore, the test method with constant confining pressure and variable internal pressure can better reflect the pressure variation properties of the actual reservoir. Many researchers have applied this method to the study of stress sensitivity (Chalmers et al., 2012; Ghanizadeh et al., 2014; Hou et al., 2017; Zeng et al., 2017). Reyes and Osisanya (2002) used the power law function to fit the relationship between porosity, permeability and net negative pressure, and obtained the stress sensitivity of porosity and permeability in unconventional reservoirs. It was found that the law of change of permeability with effective stress satisfies the exponential relation. Alramahi and Sundberg (2012) predicted the stress-related conductivity of hydraulic fractures and studied the relationship between rock mineralogy and mechanical properties of proppant embedding. Ghanizadeh et al. (2014) found that the anisotropy and stress sensitivity of permeability is related to mineral composition. Kumar et al. (2015) compared the stress sensitivity of fracture aperture and permeability with and without proppant. However, the macroscopic experiment can only compute stress-sensitive macroscopic results, with less concentration on the details of internal seepage space changes.
