Changes in Optimization Settings may cause break in old input files (Plus New Feature) #298
GabrielSoto-INL
announced in
Releases and Patches
Replies: 1 comment
-
Thanks for the heads up on this change, Gabe. I'll pass the word along to the guys at USU. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello everyone!
I wanted to bring attention to a new feature for HERON input scripts that could potentially break previous scripts when using optimization settings.
Changes to Optimization Settings
Previously, when running a HERON case in
opt
mode, a user could specify<optimization_settings>
within the<Case>
node. An optional entry within those settings was<metric>
which allowed the user to select a statistic (e.g., "expectedValue", "valueAtRisk") applied to the set of NPV results from the inner runs. This is then used as the Outer objective function.For an old example of how this was used, see here (an old version of the optimization_settings test).
We have recently switched to a format that differentiates between a statistical metric and economic metric to optimize. The old
<metric>
node is now called<stats_metric>
. Additionally, we now allow for an<opt_metric>
to be chosen besides NPV. HERON currently supports:Levelized Cost (implemented as the
NPV_search
option in TEAL) is coming soon! The new format looks like this:New Feature for Economic Metrics
There is also a new node in the
<Case>
nodes for returning additional economic metrics at the end of your run, regardless ofopt
orsweep
mode.These (along with all statistics in
<return_statistics>
applied) are reported back in the final CSV file at the end of the run.Beta Was this translation helpful? Give feedback.
All reactions