Replicate FLOPS VAPPR behaviour

[cmbenne3]

Desired capability or behavior.

Aviary should add functionality to replicate the FLOPS VAPPR behavior.

This is an input variable in FLOPS which allows the user to add an upper bound constraint on the value of output VAPP. VAPP appears in the Objectives/Variables/Constraints Summary section of FLOPS output file:

VAPP most closely maps to Mission.Landing.INITIAL_VELOCITY.

Suggested implementation.

Aviary could replicate the VAPPR behavior by adding a new constraint variable to the hierarchy (e.g. Mission.Constraints.MAX_APPROACH_VELOCITY), and then add a constraint:

Mission.Landing.INITIAL_VELOCITY <= Mission.Constraints.MAX_APPROACH_VELOCITY

When the value of Mission.Constraints.MAX_APPROACH_VELOCITY is greater than zero.

The FLOPS input variable AERIN.VAPPR should be mapped to this new variable (Mission.Constraints.MAX_APPROACH_VELOCITY) and handled by fortran_to_aviary conversion function.
An investigation should be conducted to see if there is an equivalent GASP variable that can also be mapped to this.
The new constraint should be made available for all problem types and equations of motion in Aviary.

[jkirk5]

I actually don't like the idea of porting this variable - the reason is that I don't want there to be a bunch of different ways to add constraints in Aviary. If we add a new variable specifically to add a constraint, we now kind of need to justify why every possible constraint is or is not in variable form, and then users have to know which constraints can be activated by the csv and which have to be added using OM syntax in their run script.

[cmbenne3]

I actually don't like the idea of porting this variable - the reason is that I don't want there to be a bunch of different ways to add constraints in Aviary. If we add a new variable specifically to add a constraint, we now kind of need to justify why every possible constraint is or is not in variable form, and then users have to know which constraints can be activated by the csv and which have to be added using OM syntax in their run script.

So I think I agree with you Jason - we need to draw a line somewhere. My counter argument to this is that we provide a fortran to aviary converter function, but that converter only converts some of the file. We list variables that have not been converted, but we don't provide any further help to the user on if they can or how they should replicate that functionality in Aviary.
We should consider whether we want to change the aircraft:fuel:ignore_fuel_capacity_constraint variable, but perhaps this is a more fundamental constraint that belongs in core aviary?

(here's a snip from the aviary/utils/test/data/converter_test_advanced_single_aisle_FLOPS.csv file - you see that VAPPR is set, and ignored and Aviary doesn't know how to interpret it, and it would be up to the user to think about what FLOPS does with that variable, then work out which Aviary variable to constrain as a result):

I have created a new issue to write documentation to "help users convert unconverted values" #1234 . We can close this issue and proceed with that approach if we think that would be useful (but i think it would be a very low priority).