Conversation
ecoon
left a comment
ecoon
left a comment
There was a problem hiding this comment.
Two minor changes, then this looks good.
| energ_loss_coeff_subweir_ = plist.get<double>("energy losses coeff submerged weir", 0.056); | ||
| energ_loss_coeff_orifice_ = plist.get<double>("energy losses coeff orifice", 0.167); | ||
|
|
||
| sw_domain_name_ = plist.get<std::string>("surface domain name", "surface"); |
There was a problem hiding this comment.
Can one of these domains be inferred from the key being evaluated? E.g, if you are evaluating "surface-pipe_drainage" then prefer to use:
sw_domain_name_ = Keys::getDomain(my_keys_.front().first);
There was a problem hiding this comment.
Thank you for your comment.
The domain is obtained using "Keys::getDomain(my_keys_.front().first)" to identify if we are in the surface or pipe network domain later in the code (line 54) when using the pipe_drain_evaluator.
Hence we need to set the default name of sw_ or pipe_ domain before we use "Keys::getDomain(my_keys_.front().first)" to obtain the current domain.
| if (domain1 == pipe_domain_name_) { | ||
| for (const auto& dep : dependencies_) { | ||
| auto domain = Keys::getDomain(dep.first); | ||
| if (pipe_domain_name_ == domain) { |
There was a problem hiding this comment.
What happens if the domain isn't pipe_domain_name? You should still set this evaluator's expectations of what the dependency's structure is.
Likely you should be making two "fac" variables, one with pipe_domain, one with surface_domain as the mesh. Then loop over each dependency and Update with the correct one.
There was a problem hiding this comment.
@dasvyat could you please help with this comment?
|
Thanks @ecoon . @nvohra0016 would you mind addressing Ethan's comments? |
|
Where is this PR? Some of these changes are probably old -- e.g. the constitutive relations CMake changes are almost certainly wrong. But I also don't want to lose the pipe network evaluator because people aren't being responsive. @jd-moulton can you ensure someone at LANL takes ownership of this work and gets it to the finish line? |
|
@nvohra0016 left LANL, and @dasvyat is currently testing these integrated models. Let's follow up with @dasvyat once he's done with the transport PR. |
4 participants
This PR adds a pipe drain evaluator modeling the flow of water through manholes between an overland flow model and an underground pipe network model. This PR is the companion of amanzi/amanzi#824