Deformation of digital images and trends of porosity versus numerical elastic properties
Kyle T Spikes, Morteza Elahi Naraghi · International Geophysical Conference, Qingdao, China, 17-20 April 2017 · 2017
Summary This work demonstrates that a multiple point simulation (MPS) method can be used to reconstruct a 3D image from 2D slices, then to deform the reconstructed image, and compute the elastic properties of the deformed images. Deformation of the images consists of altering the porosity of the image to mimic compaction, cementation, or dissolution. When the deformation to a given porosity value is done in a stochastic sense, multiple images with the same porosity become available. Accordingly, an estimate of the elastic properties comes from each deformed image. In turn, uncertainty estimates become available, which is not common in computations of rock properties. Trends of the elastic properties as a function of porosity reveal potentially very different behaviors compared to analytical solutions for rock physics models and bounds. Although the behavior of the trends is not yet explained in full, they might be indicative of relationships between the microstructure of the sample and the elastic properties.