Geoscience improvements: please vote!

[WilkAndy]

Check these boxes if you have followed the posting rules.

• Q&A General is the most appropriate section for my question
• I have consulted the posting Guidelines on the Discussions front page
• I have searched the Discussions forum and my question has not been asked before
• I have searched the MOOSE website and the documentation does not answer my question
• I have formatted my post following the posting guidelines (screenshots as a last resort, triple back quotes around pasted text)

Question

This week, at the MOOSE International Workshop a bunch of geoscientists met face-to-face and virtually to discuss improvements to the geoscience capability of MOOSE. Below, I've tried to list all the improvements mentioned. If you want to add something, please do!

Please vote on your favourites. Give yourself a small handful of "thumbs up" and vote on which improvements you like. This will allow developers to optimally plan.

[WilkAndy]

MISCELLANEOUS "QUICK-FIX" IMPROVEMENTS

[WilkAndy]

Need more PorousFlow examples

[WilkAndy]

Mention on the main website that the PorousFlow tests contain lots more examples

[WilkAndy]

Difficult to set up all the DiracKernels, etc, for multicomponent, multiphase problems: you need so many of the things!

[cpgr]

Ability to do 3-phase (liquid-gas-solid) modelling @cpgr : do we have this already?

Not really at the level that would be useful

[GiudGiud]

For the convergence criterion that change with time this is reasonable to do with the Convergence system

For automated setup of complex problems you could consider writing an Action / a Physics (inherit PhysicsBase and see the additional APIs there)

[WilkAndy]

COMMUNITY

[WilkAndy]

Make PorousFlow more attractive to new users

• Perhaps create a GUI for model construction
• More users would therefore help with code reviews
• More examples and tutorials would be useful
  • particularly ones including solid mechanics (eg damage)
  • guidelines for constitutive models
  • guidelines regarding convergence problems, convergence criteria, convergence tolerances, variable scaling

[WilkAndy]

Create a "geoscience virtual test bed", which would help with fully-functional examples

[WilkAndy]

Publish comparisons with other codes

[WilkAndy]

Encourage contributions
Instead of just creating our own MOOSE-Apps, can we contribute to PorousFlow (etc)?

Barriers:

• worries that no other MOOSE users are actually interested in my contributions
• SQA requirements
• need more code reviewers, because framework people don't have expertise
• difficult to create useful tests that have small runtimes and meshes
  Incentives:
• publication
• MOOSE framework developers keep stuff up-to-date

[WilkAndy]

Here is a comment from JoseB:

I've been thinking about how to ensure PorousFlow remains a lasting 
project rather than one that relies on the coding skills of just a few 
people. If we want to see it grow over time, I believe the number of 
contributors has to increase. The user base is key, as there's a clear 
correlation between the number of users, the ability to solve questions 
in the MOOSE discussion forum, and the overall development of the code.

ModFlow is, in my opinion, the perfect example of the direction 
PorousFlow should take. Without a GUI, ModFlow is quite difficult to 
master and prone to errors. However, with the addition of Python APIs 
and user-friendly interfaces, it became the go-to code for 
hydrogeologists. I think we can follow a similar path for 
PorousFlow—facilitating adoption through a well-designed GUI. User 
experience is crucial here.

Additionally, from my experience lecturing across Europe, I’ve noticed 
that young students are hesitant to learn a new code unless it offers a 
clear career advantage. Positioning PorousFlow as an easy-to-learn tool 
could help attract more users.

Of course, this involves significant development, but perhaps a hybrid 
business model—similar to what Kitware does—could support this effort. 
Providing high-quality open-source software while also offering 
consulting and custom development services might be a sustainable way 
forward.```

[WilkAndy]

GENERAL MOOSE IMPROVEMENTS

[WilkAndy]

Include specialised elements in MOOSE, such as tri6, edge3, embeddedbeam, geogrid, connector (shell-shell, shell-beam)

[WilkAndy]

Ability to include "stages" in input file for complicated time-dependent engineering simulations. Eg, https://github.com/jmeier/moose-codeplugin-stages

[WilkAndy]

Ability to plugin parts of MooseApps to other MooseApps. Eg, https://github.com/jmeier/mooseapp-code-plugin

[WilkAndy]

Parameterisation and varying parameters

1. Easy calibration. Often, we have a few observations, and want to figure out the best (set of?) material properties like permeability. Perhaps using adjoints or PEST?
2. Want ability to Monte-Carlo our parameters. Also, ability to Monte-Carlo our geometry

[WilkAndy]

Parameterisation and varying parameters

1. Easy calibration.  Often, we have a few observations, and want to figure out the best (set of?) material properties like permeability.  Perhaps using adjoints or PEST?

2. Want ability to Monte-Carlo our parameters.  Also, ability to Monte-Carlo our geometry

Does @lynnmunday agree that (1) can be done using Optimization, and (2) can be done using StochasticTools?

[WilkAndy]

WELLBORE FUNCTIONALITY

Improve handling of wellbores
Would like (multiphase) flow (NS maybe THM) inside the wellbore
Would like to couple this (potentially through casing perforations) to PorousFlow in the reservoir

[rpodgorney]

We have been doing this--will try to put an example together this spring

[Bearunner]

We have been doing this--will try to put an example together this spring

Sounds great, Robert! I am looking forward to your examples. Thanks

[Bearunner]

We plan to integrate our MOSKITO work into MOOSE. The detailed plan is being prepared.

[WilkAndy]

GEOCHEMISTRY ENHANCEMENTS

• transport of Geochem + PorousFlow to be better explained and be simplified
• need to improve accuracy for very saline groundwaters
• need to improve accuracy for high tempnerature
• terminology needs to be better explained for non-experts
• improve error messaging during the solve

[WilkAndy]

IMPROVED SOLID MECHANICS

[WilkAndy]

Improve large deformations

• This often occurs in soil mechanics and inside/around fractures. Currently get poor convergence and it's difficult to understand

[WilkAndy]

Tidy and enhance solid-mechanics plasticity + damage
There are two facets here:

1. It's difficult to understand which version of plasticity should be used: it seems a consolidation would be useful to users
2. Current plasticity models are sometimes too inflexible to use effectively, Eg, we'd like to have spatially-dependent plastic parameters

[WilkAndy]

Add more constitutive, plasticity + damage models

• hardening soil
• construction concrete

[WilkAndy]

Ability to do structural mechanics

[WilkAndy]

FRACTURES

The meeting on March 11 decided to focus on this, in particular the non-growing fracture case. Here are some notes:

• fractures will typically have O(1mm) aperture, be planar with O(10m) extent, sitting in O(1km) model
• fractures will typically form a network with complicated geometry
• we don't really care whether fractures are meshed as 2D elements, 3D elements, or not even meshed at all, but we would like the following functionality:
• Fluid and heat flow through the fractures
• Fluid flow in the bulk
• Ability to turn on/off fluid/heat transfer between bulk and fracture (which may be non-trivial since fluid/heat that "lives at" nodes that are shared by fracture and bulk may not be properly attributed to fracture and bulk)
• Geochemistry in fracture and bulk
• Solid mechanics in fracture and bulk. So, external stresses should be able to open/close (but not over-close!) and shear fractures
• Fluid and heat in fracture and bulk can change fracture aperture (which will change fracture permeability) so need to take into consideration of stresses in bulk
• Geochemistry in fracture can change fracture aperture, viz dissolution/precipitation

Fracture initiation and propagation (eg in hydrofracturing or electrical fracking) may then be possible