Determining mechanical properties of reservoir rocks through mechanical testing is a well-established procedure. Static properties obtained from these tests (Young's modulus, Poisson's ratio, UCS, etc.) are fundamental to calibrate both 1D geomechanical models (well-scale) and complex 3D models (reservoir-scale). These mechanical lab tests have the disadvantage of being destructive, making it impossible to obtain large amounts of data and limiting the generation of geomechanical reservoir models. In this context, numerical simulations of laboratory tests can be a good alternative to obtain mechanical properties of rock without the expensive destructive testing.
In this paper, we present a methodology based on numerical simulations and image analysis (thin sections and/or micrographs of cores) to model the mechanical properties from Brazilian tests simulations. The developed methodology reproduces in the simulator the mineralogy, the texture and the structure of the rock to improve the understanding of the distribution of the mechanical properties of the formations. The results are in good agreement with mechanical lab tests.
In the last few decades, unconventional reservoirs have been intensively developed after proving its economic feasibility. Among them, tight sands and shale formations have been exploited.
Those reservoirs have the characteristic of having very low matrix permeability (well below 1mD), and need to be massively fractured to generate high conductivity channels, to connect preexisting fractures to the well and/or facilitate hydrocarbon flow from the rock matrix.
The performance of those wells depends on the quality of the intended hydraulic fracture, which behavior is strongly affected by the mechanical properties of the rock. For that reason, traditional mechanical tests, such as the Brazilian, tri-axial, etc., have become standard in the oil and gas industry, to characterize reservoir rocks mechanical properties .