Interrest in multiphase flow in porous media

[Jean Francois Leon]

Hi All

I have been using the framework dumux.org  based on dune to perform multiphase heat poroelasticity fully coupled simulations [ my applicaiton is in drying green body before sintering].

It works well and its source code is available as well

its downside from my perspective are

-(very) difficult to run in parallel  and 3d "real" cases need parallel.

-focused on geophysic application [ impact the grid choices..]

On the other hand I have seen One publication from INL ( in 2011) mentioning an app "Meercat" that could be doing that.

There is difficulty along the way as it is notorious that fully implicit solving for these equations can be VERY difficult to achieve...but maybe a multi app approach will be the way to go then....

Anyway this email is really to ask:

Is there any work currently going on in the public sphere in developing multi phases flow in porous media ( with or without chemical reactions)? ( I noticed thatt Moose already have IAWS...)

Is there anybody out there in the Moose community that is or could be interested in such a subject ? 

thanks

JF

[Andrew....@csiro.au]

Hi Jean,

 

Let me tell you what i’ve done, and what i know about other people’s work.  I would love it if you built on my work.

 

(1) I’ve coded the multi-phase Richards equation into MOOSE, and it is being used extensively at my institution primarily to simulate underground fluid flows around coal mining operations (water and methane gas).  The doco is:

 

http://mooseframework.org/wiki/PhysicsModules/Richards/

 

Note there is no phase transition here: it is just a certain number of phases that do not transform into each other, but affect each other through capillarity.  This is standard for solving water+gas, or water+oil+gas, or water-only, etc type of systems.  However, other people have been solving water+steam problems using MOOSE – i’ll let them chime in here.  Also, Rob Podgorney and collaborators at INL have used MOOSE to study liquidwater+heat systems in geothermal environments – i think their moose application is not yet open-source.

 

(2) I’ve coded poro-mechanics for fully-saturated single-phase Darcy flow into MOOSE.  This is recent work, and some doco is at:

 

http://mooseframework.org/wiki/PhysicsModules/TensorMechanics/TensorMechanicsBasics/PoroMechanics/

 

The examples given in that page are just poro-elasticity, but that is just because of the availability of analytical solutions: the formulation also works for poro-elasto-plasticity, and this is being used by other colleagues at CSIRO.  Also, there are no examples yet for poro-thermo-mechanics, but i believe that MOOSE should have that capability as it stands.  I’m trying to get other people to do some examples/tests of this feature since i’m currently working on other things.  Please feel free to chip in!  The next step I would love to do would be poro-mechanics for unsaturated/multiphase Richards flow.  The equations are documented in that link above, but i just haven’t had time to implement them yet, and there are some architectural choices regarding the evolution of porosity that I haven’t figured out yet.

 

 

Other things I know about:

 

(1) Thomas Poulet and collaborators have an application called ‘redback’ that does thermal-fluid-mechanical-chemical couplings in certain situations.  They appear a little reticent to contribute this into a module, but i’m pretty sure they’d love you to use their code.

 

(2) Chris Green is extending the multi-phase richards stuff to multi-component fluids (eg, mixtures of CO2 and CH4), but i think he’s in the testing phase and he hasn’t yet contributed that work to modules.

 

a

 

 

 

 

Ph: +61 7 3327 4497.  Fax: +61 7 3327 4666
Queensland Centre for Advanced Technologies
PO Box 883, Kenmore, Qld, 4069 
 

--
You received this message because you are subscribed to the Google Groups "moose-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to moose-users...@googlegroups.com.
Visit this group at http://groups.google.com/group/moose-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/8011a8e6-0796-46f2-8655-b7b744bcd5ea%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Thomas...@csiro.au]

Hi Jean,

Andy is correct, we’re working on redback as a separate application for now since we’re still in the phase where we’re trying to figure out what we’re doing… :-)

It’s all open though so feel free to have a look if you want (https://github.com/pou036/redback). Note however that it’s really early in its development, in dimensionless form, and until we have some proper documentation and papers out to explain how it works, it will be difficult to use as is. As such it’s still very premature to think of it as a moose module, but if anybody else finds any bits of it useful later on (which we hope), then we’ll see how to port it to the poromechanics framework Andy is doing so well.

Cheers,

Thomas.

[Jean Francois Leon]

Hi Andy

Thanks for sharing your insight/info.

 I will dig into that into details

and definitively look and play with Richards Equation. 

Phase change can carry away very large amount of energy and can be the source of "difficult to model" effect ( critical flux and boiling curve- heat pipe effect for eg

Another really important piece of coding when you do multiphase calculation is the relativity permittivity /capillary pressure business.

 I see you have done some stuff in the Richards equation module but I still have to look at it ( I am getting used to Moose  and discovering its community at the moment...)

A general newbie  question in that context:\

 is it OK in this list to discuss application specifics issue such as  plans or projects or is this list intent solely focused on Moose code devt? 

( in that case I apologize for the disturbance)

cheers

JF

[Jean Francois Leon]

Hi Thomas

 thanks for the insights

I am going to have a look at this animal. It seems it might have some stuff I could use

Cheers

JF

[Cody Permann]

We encourage application discussions on the list. This policy helps drive the community development of the physics modules which is where all of Andy's contributions have been added.

Cody 

--
You received this message because you are subscribed to the Google Groups "moose-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to moose-users...@googlegroups.com.
Visit this group at http://groups.google.com/group/moose-users.

To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/6577ef80-8492-4a02-aeb5-b5a0bb88c421%40googlegroups.com.

[Yidong Xia]

Hi Jean,

This is Yidong Xia from INL's Energy Resource Recovery & Sustainability Department. I work with our department's lead Robert Podgorney, and our subsurface geosciences group lead Hai Huang. Our department has been consistently interested in the research, development, and improvement of subsurface geosciences and technologies at all spatial and temporal scales (computationally & experimentally).

In addition to Andrew Wilkins' and Thomas Poulet's works, I would like to share some basic information with you on our department's FALCON code, and its recent progress.

The FALCON code, which stands for "Fracturing And Liquid CONvection", is a highly parallel, finite-element geothermal simulator built on top of the MOOSE framework. 

• FALCON has been designed to solve the coupled thermo-hydro-mechanical-chemical (T-H-M-C) processes in geological porous media at reservoir scale. It provides two flow and heat transport models in porous media: 1) pressure and temperature based single-phase (water) model, and 2) pressure and enthalpy based two-phase (water-steam) model. We use the IAWS-97 for material property calculations. 
• By taking advantage of the MOOSE framework, FALCON can solve T-H-M-C problems either sequentially through each of its four components, or in a globally implicit manner, though the latter is known to have much greater difficulty in convergence, as you also have noted.
• We are now almost close to the release of FACLON as an open-source code on GitHub (hopefully within this July). The open-source version will initially feature the single-phase T-H capability in porous media. As FALCON will go open-source from then on, we are paying more attention on the extensibility and reusability of our source code, because we want to sparkle more opportunities of external collaboration. That is why we choose to gradually move more capabilities into the open-source version, as the revision of our in-house version goes on).    
  

Below is a not-so-recent paper that introduces FALCON. We hope to hear back from you for further discussion on potential collaborative work.

Podgorney, R., Hai. Huang and D. Gaston (2010), A Fully-Coupled, Implicit, Finite Element Model for Simultaneously Solving Multiphase Fluid Flow, Heat Transport, and Rock Deformation, Geothermal Resources Council Transactions, 34: 427-432.
http://pubs.geothermal-library.org/lib/grc/1028682.pdf

Best,

Yidong Xia

yidon...@inl.gov