Minimal corrections to Phasor Dynamics models for validation - #538
Conversation
|
Is this meant to supersede #525 ? |
Yes, I forgot to mention that |
nkoukpaizan
changed the title
nkoukpaizan
left a comment
nkoukpaizan
left a comment
There was a problem hiding this comment.
A few comments. Otherwise looks good!
abirchfield
left a comment
abirchfield
left a comment
There was a problem hiding this comment.
A good compromise that addresses the issues raised in #525 while allowing us to have excellent validation in the short term. Exciting that we are almost ready to complete the 10k case.
dbbafb2 to
2f1ea87
Compare
|
added specification information and warnings everywhere |
nkoukpaizan
left a comment
nkoukpaizan
left a comment
There was a problem hiding this comment.
I recommend setting Ke_eff_ uniformly in setDerivedParameters and moving the warnings there (instead of in initialize).
6881eff to
c5b70ae
Compare
|
@pelesh @nkoukpaizan Let me know if you think this is ready! Good on my end |
Fine by me... I can always find something to nitpick, but I'll stop there. I'll let @pelesh have the last word. |
pelesh
left a comment
pelesh
left a comment
There was a problem hiding this comment.
All major issues we discussed were addressed and all tests pass.
There is a couple of nitpicking issues that would be good to address before we merge.
| auto check = [&](bool condition, const char* message) | ||
| { | ||
| if (!condition) | ||
| { | ||
| Log::error() << "Esdc1a: " << message << '\n'; | ||
| ret += 1; | ||
| } | ||
| }; |
There was a problem hiding this comment.
this wasn't added in this pull request, but as i have said in #460, this should 1) not increment an error count---this is a bad pattern and 2) should be a common utility function, not something we define in each case it is used. this comment applies to all instances of this closure, not just this one instance
5 participants
Description
Validated against PowerWorld. Less than 0.5% relative error on WECC 240 case, ACTIVSg200 case, and ACTIVSg500 case.
Further comments
Other open-source grid modeling frameworks do not even come close to this low a validation error against their validation tools, especially not for cases this large. I don't want perfect to be the enemy of good + we can do cleanup more confidently after merging #461.