WP1.2 Coordination Meeting April 21, 2020

Meeting Report WP1.2 ‘Modelica library for MPC’

1. MEETING SUBJECT, DATE

Subject: WP1.2

Date: 21-04-2020

Location: Skype for Business

Minutes taken by: Lieve Helsen (KU Leuven)

2. PARTICIPANTS

Company/Organisation - Participant

KU Leuven - Lieve Helsen (WP Leader)

KU Leuven - Iago Cupeiro

KU Leuven - Javier Arroyo

LBNL - David Blum

LBNL - Michael Wetter

SINTEF - Harald Taxt Walnum

PNNL - Sen Huang

SDU - Toa yang

SDU - Konstantin

Oak Ridge national Lab – ORNL (US) - Yeonjin Bae

3. AGENDA and REPORT

3.1. Approval/Modification Agenda

Approved

3.2. Status & further discussion where needed

a. Emulators:

4 models selected to start MPC testing

i. BESTEST Air (LBNL/KU Leuven):

In pull request (Dave).

Dave remarks that he has been cleaning up unit tests and is providing a solution to specify the dependency libraries of each test case. Additionally there is development on starting simulations with a warm up period at time different from zero. To be reviewed by Javier.

ii. BESTEST Hydronic (KU Leuven /LBNL):

In pull request (Filip).

Unit testing: similar status as BESTEST Air.

iii. 8-Zone Residential Hydronic (Engie/SDU):

In pull request, update needed for zone identifiers (Javier).

iv. Multi-Zone Office Hybrid Simple (KU Leuven/Engie):

Managed to compile the model with JModelica, still some issues in BOPTEST (action Iago/Javier). Check with Sen. Simulation time in FMU could be much longer than in Dymola. Changing solver might help.

However further progress has been made in other models too!

v. Single-Zone Commercial Building (SDU/LBNL):

Dave gave feedback and this is processed by Tao.

vi. Multi-Zone Air-Based Prototype Complex (PNNL/KU Leuven):

Iago gave feedback, occupancy profile is missing. To be processed by Sen.

vii. Flexible Research Platform (ORNL/SDU):

Yeonjin started with the building envelope model for the 10 zones office with VAV system. Konstantin volunteers as a Reviewer. Dave sends the checklist.

Emulators need to be put on BOP-TEST repository in the directory Testcases

Emulators overview:

b. KPIs - scenarios

Choice between TMYx (1946-2018) and TMY3 (1973-2005) weather files. We have a data reader for TMY3 in IBPSA. Both will be compared in the representative days/weeks selection (action PNNL). Sen will give an update next meeting.

c. BOPTEST:

i. Test case peer review document:

Unit test for the functionality of the zone name designation to the measurement read block has already been merged with the BOPTEST master (Dave/Javier).

ii. Qi Li (no longer at ORNL, moved to Siemens)

He sent a paper to Dave on the generation of new weather files based on historical data, to be used for grey and black box model identification (training data that differ from test data). No further progress.

iii. Dave reports from the Technical Advisory Group meeting for US part:

Good participation (13 attendees), Dave presented the test case construction and emulators, feedback was positive: emulators cover a wide range. Comments:

1. Add faults to the system and see how controllers respond.

2. Push towards the latest efforst on standardization of data (I/O) tags ... the process of discovering data points easier, like Project Haystack, Brick Schema, that are initiatives from the ASHRAE Standard 223P. Could BOPTEST interface with these standardization efforts?

3. Important to include documentation, e.g. control sequences included in emulators, incl. HVAC schematics and diagrams.

d. MPC Library (IbpsaMpc):

i. Unit testing framework (Dave):

No update.

e. Dissemination:

Broaden the user space (e.g. to Machine Learning people). How to reach these people? Jan: make a first suggestion. Cfr OpenAIGym, no update.

3.3. Initiatives joint papers

a. MPC Review paper: Jan was going to submit.

3.4. Next phase

a. Finalize first set of emulators by next progress meeting.

b. Start testing your controller(s) on the emulators and share your comments, remarks, suggestions for improvement.

c. In the meantime the development of the remaining emulators can start.

3.5. Miscellaneous

a. Break-out sessions definition for Montreal postponed (Lieve)

i. Spreadsheet for internal reporting of BOPTEST test cases

ii. Public reporting (limited information, end-user agreement)

iii. MPC competition: how to set up?

iv. How to deal with uncertainties: forecast of weather, loads, occupants?

b. Montreal meeting postponed to October (probably the week of October 12).

c. May 6: Plenary IBPSA Project 1 meeting (agenda on Wiki), Progress to be presented (Lieve).

d. Next progress meeting: June 1 (invitation to be sent, Lieve)

e. June 18 Planning meeting Montréal for WP Leaders (Lieve)

f. IEA-EBC Annex 81 has kicked-off, lots of interest in MPC (algorithms) comparison, particularly from the data-driven side. They want to use BOPTEST. Dave is in.