2020 11 18_meeting - lbl-srg/soep GitHub Wiki

Date: November 18, 2020, 8:15am-9am Pacific Times

The purpose of this meeting is to coordinate the Modelica integration for SOEP.

Join from PC, Mac, Linux, iOS or Android: https://lbnl.zoom.us/my/mwetter

Or iPhone one-tap (US Toll): +14086380968,6614042296# or +16465588656,6614042296#

Or Telephone:

Dial: +1 408 638 0968 (US Toll) or +1 646 558 8656 (US Toll) Meeting ID: 661 404 2296 International numbers available: https://lbnl.zoom.us/zoomconference?m=_h5BuJ686mPy3rWEyKf4NROxLaeOV5J8

1. Previous action items.
2. Release of Spawn coupling.
3. Kyle, Michael: Refactoring room air heat and mass balance.
4. Kyle, Michael: EnergyPlus I/O variables.
5. Michael, Kyle: Support for radiant systems.
6. Stuart: FMI-QSS.
7. Kyle, Michael: Optimica license protection.
8. Jianjun: JSON representation for OS front-end to Modelica
9. Other items

See milestones.

Issues are at https://github.com/lbl-srg/modelica-buildings/issues?q=is%3Aissue+is%3Aopen+label%3Aspawn

Next, work on

• Spawn FMUs are not yet self contained, https://github.com/lbl-srg/modelica-buildings/issues/2054

• [For FY21] Kyle to fix schedule: https://github.com/lbl-srg/modelica-buildings/issues/2000

  (Kyle either fixes it, or proposes new wording for https://lbl-srg.github.io/soep/softwareArchitecture.html#sending-input-to-energyplus)

• [For FY21] Kyle to look into numerical instability (walls may be initialized to wrong temperature). See for example Experimental/EnergyPlus/Examples/VAVReheatRefBldgSmallOffice/FloorOpenLoop. The temperature in the attic should not raise in the first hour. (Pending since Nov. 2019)
• [Done, except for correction of unit strings] Kyle to provide list of supported output variables, and statement of how non-supported output variables are handled, https://github.com/NREL/Spawn/commit/e443d634ab4f15d5d69e487fcc206c7d4f403353 Outputs not in idf but supported will work. Outputs not supported just won't be in the xml file, Kyle will work on issuing a log when spawn.exe is called with an output variable that does not exist. Actuators are handled differently, may be addressed later.
• Michael/Milica to use thermal zones in larger model, https://github.com/lbl-srg/modelica-buildings/issues/1502
• [Completed] Kyle to move model.spawn to resources folder. It has the same information as ModelicaBuildingsEnergyPlus.json (other than that some path are absolute and some are relative in model.spawn). https://github.com/lbl-srg/modelica-buildings/issues/2215. Binaries are posted.

LBL AOP 9/30 (go/no-go): Release that can couple Modelica-based control to E+.

Kyle to work on emitting all EnergyPlus objects that are supported by Spawn.

1. If there is a thermal zone in Modelica, then there need to be a Zone object in the idf file. Otherwise the simulation stops with an error. EnergyPlus reports a Fatal Error to its error file and to the FMI API call.
2. If there is an Zone object in the idf file but no thermal zone in Modelica for this EnergyPlus zone, then EnergyPlus simulates the zone using free floating temperature. There is no HVAC system, but may be internal gains. EnergyPlus reports an information message to its error file and to the FMI API call.
3. All HVAC objects in the idf file will be removed by the Spawn executable before it writes the idf file to the FMU.
4. For the EnergyPlus envelope, either the CTF transfer function or the finite difference method can be used.
5. Output variables and EMS actuators need not be present in the idf file.

Specfication at https://lbl-srg.github.io/soep/softwareArchitecture.html#coupling-of-a-zone-surface.

To be done in FY 21.

Stuart works on estimate for how much function calls are reduced if there is no numerical noise from derivative approximation.

Stuart works on benchmarking of events.

No discussions.

Kyle to look into license protection provided by Modelon.

No discussions.

To do:

• [closed] Address TypeError issues reported in https://github.com/lbl-srg/modelica-json/issues/106.
• Handling replaceable models.

None.

None.

Michael to keep working with Modelon on making dp of pressure drop elements a state rather than p using stateSelect. Check tearing variables and make sure they don't go to zero.

Kyle to look into license protection provided by Modelon.

Michael to test on OpenModelica.

Michael to make new issue about logging of schedule and rel. humidity ( https://github.com/lbl-srg/modelica-buildings/issues/2220)