Estimation of 3-D pore network coordination number of rocks from watershed segmentation of a single 2-D image

Rabbani, A ; Sharif University of Technology | 2016

219 Viewed
  1. Type of Document: Article
  2. DOI: 10.1016/j.advwatres.2016.05.020
  3. Publisher: Elsevier Ltd , 2016
  4. Abstract:
  5. In this study, we have utilized 3-D micro-tomography images of real and synthetic rocks to introduce two mathematical correlations which estimate the distribution parameters of 3-D coordination number using a single 2-D cross-sectional image. By applying a watershed segmentation algorithm, it is found that the distribution of 3-D coordination number is acceptably predictable by statistical analysis of the network extracted from 2-D images. In this study, we have utilized 25 volumetric images of rocks in order to propose two mathematical formulas. These formulas aim to approximate the average and standard deviation of coordination number in 3-D pore networks. Then, the formulas are applied for five independent test samples to evaluate the reliability. Finally, pore network flow modeling is used to find the error of absolute permeability prediction using estimated and measured coordination numbers. Results show that the 2-D images are considerably informative about the 3-D network of the rocks and can be utilized to approximate the 3-D connectivity of the porous spaces with determination coefficient of about 0.85 that seems to be acceptable considering the variety of the studied samples. © 2016 Elsevier Ltd
  6. Keywords:
  7. Coordination number ; Pore network extraction ; Pore network modeling ; Estimation ; Rocks ; Determination coefficients ; Distribution parameters ; Mathematical correlation ; Pore networks ; Pore-network modeling ; Watershed segmentation ; Watershed segmentation algorithm ; Image segmentation
  8. Source: Advances in Water Resources ; Volume 94 , 2016 , Pages 264-277 ; 03091708 (ISSN)
  9. URL: http://www.sciencedirect.com/science/article/pii/S0309170816301543