Skip to content

Monolithic coupling WIP#10

Open
kmokstad wants to merge 23 commits intoOPM:masterfrom
kmokstad:monolithic
Open

Monolithic coupling WIP#10
kmokstad wants to merge 23 commits intoOPM:masterfrom
kmokstad:monolithic

Conversation

@kmokstad
Copy link
Contributor

@kmokstad kmokstad commented Jul 8, 2016

This branch contains items for the monolithic coupling. Work-in-progress, so no merge yet.

@kmokstad kmokstad force-pushed the monolithic branch 2 times, most recently from 3cf0ebf to e53202b Compare July 11, 2016 09:22
@kmokstad
Copy link
Contributor Author

It compiles and builds, but does not work yet. Requires OPM/IFEM#36.

akva2 and others added 14 commits September 10, 2018 06:57
- use self-written hermitian interpolator (GoTools is buggy)
- find zeros of the derivative using a Newton loop instead of
  a (naive) line search
Added: Input of number of (uniform) points, or explicit point parameters,
and tolerances for the line search. Added search from all grid points.
…search.

Fixed: Sign of the derivative function f'(alpha).
Added: Option to dump out f(alpha) and f'(alpha) to file.
Fixed: There was also an error in the calculation of f(alpha) itself
(missing Gc-factor in front of the gradient term).
before checking convergence. Not sure if this will help though.
Using L2-norm of the residual force vector for convergence check.
staggered solver, such that input files with initial condition setup for the
monolithic version also can be used by the staggered solver without issues
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants