Any recommendations about this topic are highly appreciated!
Thanks!
By studying the relations between fracture properties and the physical properties you have access to. One you have those relations (or interpretation of those relations) you can generate fracture sets mathematically conditioned to be within the expected values. This is particularly true for stochastic based studies (which later you can use for history matching for example). Take a look at this:
https://www.researchgate.net/publication/215733644_Natural_Fracture_Networks
The document above hypothesizes a possible mathematical conditioning (for fractures) for a real reservoir in the southeast of Arabia peninsula where porosity (high) mostly comes from chemical substitution.
You can also condition your fracture estimation with geomechanics but this is mostly useful for the bigger families.
Thesis Natural Fracture Networks
For dealing with fractures and high porosity, you must have nonlocality (nonlocal kernel) in your constitutive relation.
In addition, some articles.
http://www.eurekalert.org/pub_releases/2006-01/m-fmt012006.php
http://www.nature.com/nature/journal/v439/n7074/abs/nature04408.html
http://www.pnas.org/content/109/2/390
http://phys.org/news/2005-11-glass.html
If I had to model fractures, I'd start with a graph.
Points are distributed randomly or systematically
in space or plane. Then Voronoi tesselation should
be applied to obtain edges (centers of directly
neighboring cells should be connected). These edges
should be local ones. Maybe there is a method to get
local edges easier without Voronoi. In a first
step you could use a 2D model, connect
points with a mouse and store these data.
The edges have some strength which must be
overcome to destroy it. And some energy, which is
set free and propagates among the remaining edges,
when an edge is destroyed. How the energy is
propagated is a matter of playing. Maybe it
is distributed proportional to the cosinuses of
the angles of the edges, maybe time dependent
as decaying and dampened wave.
Regards,
Joachim
One approach is to learn from skeletonization and thinning algorithms used in image processing that is in the area of segmentation and morphological studies. If you are interested in the visualization aspects you can try ImageJ' s boneJ and others such as Mathematica's Skeletonize function.
Dear Mohamed Ramy El-Maarry
Please look at the articles below:
https://www.researchgate.net/publication/269097674_ULTRASOUND_PROPAGATION_IN_FRACTURES_WITH_INTRAMEDULLARY_NAILING
https://www.researchgate.net/publication/236664175_Computational_simulations_of_time_of_flight_and_attenuation_of_first_arriving_signal_from_healing_process_of_diaphyseal_femur_fractures
https://www.researchgate.net/publication/259878491_The_effect_of_low-intensity_therapeutic_ultrasound_in_induced_fracture_of_rat_tibiae
Regards,
Conference Paper ULTRASOUND PROPAGATION IN FRACTURES WITH INTRAMEDULLARY NAILING
Article Computational simulations of time of flight and attenuation ...
Article The effect of low-intensity therapeutic ultrasound in induce...
Salam. As I know "finit element " modeling, is one of the best modelling apparatus for bone changes, specially pathologic fractures. Bye and good luck.
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling