Skip to content

Bound and variable information for PET model comparison #222

Open
@pauldhein

Description

@pauldhein

Planning

We need to be able to demo how SA can be useful to comparing two models that claim to be modeling the same phenomena. To do this we will need bound information as well as optimal variable setting information for the PT and ASCE models.

Todo's

  • @skhan1020 has already identified a good range of input space for the PT model. @skhan1020 please do this for the input variables to the ASCE model as well.
  • @skhan1020 given the two FIBs shown below for the overlap between the PT and ASCE module please identify good present values for all of the green nodes. These nodes are nodes whose values will need to be set (because we must observe these nodes) to process the forward computation through the FIB model
  • @skhan1020 after these are done we may also need to define subranges for all the mutual inputs to the PT and ASCE module that will be most interesting for comparison purposes. However let's hold off on this for now.

FIB Diagrams

PETPT FIB

pt-fib

PETASCE FIB

asce-fib

Metadata

Metadata

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions