Use mblv9

[vtnate]

Any background context you want to provide?

The GMT should work with mainline MBL, to enable new capabilities as they are developed

Until we have a compiler for MSLv4, success for this PR will be defined as:

• the manual test procedure working no matter which model you open
• all other pytests passing (as of 2022-05-11 this is 18 failing, 78 passing)

What does this PR accomplish?

Update templates, code, and test files to work with MBLv9

How should this be manually tested?

• Check out MBL master branch
• poetry run py.test
• open a model in Dymola
• Simulate the model
• Bask in the glow of a simulated model without errors 🎉

What are the relevant tickets?

#409

[vtnate]

@nllong The failing test has a Poetry deprecation warning. I don't know if that is related to the check failure or not. Can you check sometime at your convenience? I don't think it's urgent.

[amyeallen1]

This looks great! Wonderful job, Nate! I ran the tests and replicated your results in terms of the # passing and failing, and ran the CHP system, district htg & clg with Teaser, district cooling (time series), and solar PV models in Dymola. They generally performed as expected, with a few specific notes below about things that are beyond the scope of this PR:

Regarding a few specific points:

• CHP model: The high HHW supply water temperatures seem to happen for short periods of time at conditions of low- or no mass flow, so I agree that it isn’t a problem. I noticed a possible area for future investigation in terms of the performance of the CHP model, and I'll submit an issue. (When I ran the simulation for a 2-month period, I noticed quite a few periods of negative mass flow at the supply temperature sensor (also corresponding to negative mass flow at the return mass flow rate sensor). )
• This branch is incorporating changes that will address the inconsistency in weather files between the CHW plant and building load models (such as Teaser) that currently occurs in the models generated by the tests: https://github.com/urbanopt/geojson-modelica-translator/tree/schema_mod_weather_file

[vtnate]

@nllong Thank you for fixing the CI workflow! I could use some education on CI at some point in my career, it continues to be a mystery to me in general.

Local tests all pass when the not msl_v4_simulation message is added to pytest. With the excellent review by @amyeallen1 I am confident that the models run as expected in Dymola.

[amyeallen1]

Aww, thanks, Nate! Wonderful job! @vtnate

[nllong]

@vtnate -- any reason not to merge once these tests pass?

[vtnate]

Nope, let's do it! 🚀