Sebastian Blauth
Sebastian Blauth
At the moment, it may happen that Newton's method diverges and produces larger and larger residuals. Users should be able to specify some (relative) divergence tolerance, so that when this...
This PR adds support for Python 3.12 for cashocs. Closes #325
At the moment, most of the default configuration is duplicated in the demo configurations. This should be changed, so that only the non-default options (so the ones that do alter...
At the moment, the parameters for the nonlinear CG methods are specified, e.g., as ``` cg_periodic_restart = True cg_periodic_its = 5 cg_relative_restart = False cg_restart_tol = 0.5 ``` So boolean...
In the config file, users should be able to specify `deformable_subdomains`, which would be a list of integers, indicating which subdomains of a given mesh are deformable. Additionally (or alternatively),...
To incorporate volume constraints in topology optimization, the level-set can be projected (shifted by a constant) to achieve the desired volume. It would be nice to implement this in cashocs,...
The newly introduced global deformation (see #222) should be used for space mapping for shape optimization (this makes the space mapping also work with remeshing).
Pseudo-Transient Continuation seems to be a robust way of globalizing Newton's method. It should be possible to implement this for general problems in cashocs. One note: Users should have the...
At the moment, topology optimization does not work in parallel.
Cashocs should be able to make more use of logging - particularly with the debug logger (which isn't enabled by default). Things such as the stepsize computation should be verbose...