2017 05 10_meeting - lbl-srg/soep GitHub Wiki

SOEP Coordination meeting

Date: May 10, 2017, 8am-9am Pacific Times

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

Meeting information

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

Agenda

  1. Previous action items.
  2. David: Setup for "minimum viable product" with an end-to-end demonstration for a 1 zone model and an idealized system.
  3. Mark: OS to Modelica translation.
  4. Edwin: Update about heat conduction implementation.
  5. Michael: Buildings API.
  6. Michael/Jianjun: Scalable benchmark.
  7. Thierry: Update about FMI-QSS.
  8. Michael O'Keefe: Documentation/Meta-data Generation Tool Support (see proposal here)
  9. Kyle: Example OS measure.
  10. Other items

Participants

Michael O'Keefe
David Blum
Jianjun Hu
Michael O'Keefe
Mark Adams
Edwin Lee
Michael Wetter

Minutes

Minimum viable product

Dave works on fan energy discrepancy for single zone VAV. Reason is the E+ has higher supply air temperature leading to twice higher fan energy consumption. Discussions with E+ help desk are ongoing, not clear based on documentation how E+ controls system.

A change to the Modelica fan model is close to completion. This change will allow avoiding algebraic loops that are caused in systems in which the fan static pressure is reset based on a measurement in the duct distribution downstream of the fan.

Heat conduction implementation

Edwin didn't have much time to work on it. Best would be to use Spitlers's periodic transfer function approach. Edwin will have more time in June, but will do some more time before that. Michael said this will be critical for simulation time benchmarking as heat conduction leads to large computing time due to O(n^3) scaling.

Buildings API

Michael W. presented two outlines for library adaption. Michael O said a 3rd approach would be a text based template that generates the .mo file.

Michael O presented JModelica-parsed AST for mover. This is well along.

Next would be to further design and discuss among Michaels and with Kyle.

Scalable benchmark model

Jianjun reports that first version is finished, has vectorized VAV model with different internal heat gain, start points and magnitudes. Tested it in Dymola and JModelica.

Now works on Python code to test translation and simulation time, then will work on QSS coupling when QSS becomes available.

FMI-QSS

Work ongoing on zero crossing and state event detection, hopes to wrap it up this week.

Documentation

Presented demo for parsing AST of Buildings library. AST parser is based on Java.

OpenStudio API

Mark needs to touch base with Kyle to design and develop parser that uses Michael O' AST.

Packaging

Last week Kyle and Michael had a talk with Modelon (Christian Winther and Hubertus Tummescheit). For now, we will need to ship Python, but this could be revisit in a year or two. Currently, we will need Python to prototype and implement the master algorithm.

General

No update.

New action items

None.

Previous action items

  1. Kyle to prepare OS measures in order to move forward with how to design OS-Modelica API.
⚠️ **GitHub.com Fallback** ⚠️