2017 08 05 kick off agenda

Michael Wetter edited this page Aug 30, 2017 · 8 revisions

Project 1 Kick-Off Meeting

Date: Saturday, August 5, 2017, 8-12.

Location: Hyatt Regency, Room Bayview B, San Francisco, CA

Registration is required here by July 30, 2017.

Who should attend? Developers and users of building and district energy simulation programs who intent to actively contribute to the IBPSA Project 1. See https://ibpsa.github.io/project1/participants.html for membership levels and https://ibpsa.github.io/project1/ for tasks and workplan.

Draft Agenda

Time Description Presenter
8:00 Introductions All
8:20 Project and team structure Michael Wetter and Christoph van Treeck
8:30 Overview of each subtask
  • Subtask 1 (15 min)
  • Subtask 2 (15 min)
  • Subtask 3, key demonstration projects (20 min)
Subtask leaders
9:20 Workplan review and committment of participants All
10:20 Update about current activities Michael Wetter and Christoph van Treeck
10:40 Dissemination strategy All
11:40 Next meetings Michael Wetter and Christoph van Treeck
11:50 Wrap-up Michael Wetter and Christoph van Treeck
12:00 Adjourn

Minutes

Project and team overview

Project 1 is a continuation of Annex 60 with a change in focus:

  • Annex 60 generated new tools based on Modelica and FMI.
  • Annex 60 final report is approved by reviewers and needs to be approved by chair.
  • Final report will be available as book and pdf on website.
  • Meetings and workshops were held previously in preparation of Project 1. The workshop was to teach Modelica and FMI with about 60 attendees.

Overview presentation: see slides posted online

Questions

  1. Question about IP in France - the code created is protected by owner rights, so how transfer to IBPSA?
  • Pull request on Github is understood that you are contributing code under the license of repository.
  • Could also have agreement with author to sign if needed.
  • Organizers may need to check up on this.

Overview of subtasks

For details, see slide presentation posted online.

Task 1) Modelica libraries for building and community energy systems** - led by Michael Wetter:

  • WP 1.1: Library for design and operation - led by Michael Wetter

    • Annex60/IBPSA Library bases of AixLab, BuildingSystems, Buildings, OpenIDEAS - but each libary customized afterwards.
    • Plan - continue and expand collaboration from Annex 1.1.
    • Talk about creating IBPSA Air and IBPSA Water in addition to fluid.
    • Work on scalability for large system models.
    • Add other models, such as borefields, absorption chiller, chp, template HVAC, control sequences.
    • Standardize representation (data, physics, controls, documentation) and distribution.
  • WP 1.2: Library and approaches for MPC - led by Lieve Helsen

    • Enable MPC for buildings and district energy at scale
    • Repository of models
    • Data streams
    • Repository of optimization algorithms
    • Key Questions:
      1. linear, piece-wise linear, non-linear?
      2. how use design model?
      3. white-box, grey-box, black-box?
      4. deterministic vs. stochastic?
      5. state and parameter identification?
      6. how encode libary to support?
      7. how to develop frameworks to support?
      8. how to benchmark? - BESTEST for MPC that covers requirements, model identifiation, MPC

Task 2) Building and City Quarter Models - led by Christoph van Treeck:

  • WP 2.1: City Quarter information modeling
  • WP 2.2: Building information modeling
  • Part of work is to determine subpackages - not as clear as WP1.
  • Digital planning and simulation - from geoinformation system and BIM to district system and building HVAC etc.
  • IFC --> Modelica.
  • Early-stage design - HVAC Line Diagrams and basic building design block diagrams.
  • Model-based commissioning and operation use.
  • Standardization of definitions and classifications. Communication problem with integral planning of components and groups of components.
  • How communicate physical components and groups to automation/control representation?
  • Below ground heat transfer modeling.
  • Functional descriptions within BIM.
  • SimQuality - Validation of existing models at building scale - with Dresden too.

Task 3) Application and Dissemination - may be led by Dirk Saelens?

  • WP 3.1: Application, including BESTEST for Urban, DESTEST?

    • Application and demonstration focus.
    • BESTEST for district energy models.
    • Case study reports and validation test procedures.
    • Papers and presentations.
  • WP 3.2: Dissemination

    • Work with individual conference organizers to set up specific tracks and programs related to IBPSA Project 1.
    • Publishing reports, papers, procedures, etc.

Questions and comments

  1. Great plan, but it needs to address the education problem at University level and companies. How can we address this? Documentation sure, but what else?

  2. As scientist:

  • BIM:
    • if go around room, argue that 60 people = 60 different definitions
    • need to agree on what BIM really needs
    • testing - really means certification
      • costly - if those who are testing are not willing to provide cost - then not going to happen
      • testing is not just that something runs - but also thorough testing of input and output
    • data
      • in this industry, data is terrible
      • no accountability

Answer:

Working with IBPSA will help, instead of IEA - specific to building simulation, etc.

Add comment:

Current gap of industry vs. modelica models - reflect models with real devices - took lots of work to go from IBPSA model to real/useable model

Answer: Focusing a lot on base classes to make sure they're robust.

Need more people contributing actual models - including manufacturers who have more experience to ensure that the models reflect what's needed.

Team up people who know products (manufacturers) with those who can develop library (students).

Add comment: Issues require more conversation:

  • straight forward questions: how represent, how define interfaces...
  • higher level questions: wish list is clear - ideally have global model of city, with optimization, etc.
  • need to be realistic on actual possibilities going from different scales - some things may be impossible to do
    • example: HVAC systems - field is evolving and developing all of the time - define interfaces instead?
    • at global level - need to see what complexities we can solve

Add comment:

In IT and internet, had problem before, solved by platform/framework called xml and queuering language to abstract. Once have structure - fill with data - and extract data.

Answers:

In modeling domain, we can use xml to transfer data from A to B, but question is how to classify systems extensively - using, designing, operating?

  • need classification to structure things on different scales and levels

Two questions on this:

  • all working at different scales on specific questions - CO2, cost, energy, etc.
  • have specific answers for certain problems - want to expand them - functional descriptions/classifications

Response:

Availability of data is 50-80% of problem, not just structure.

  • Who would like to contribute on community modeling and transformation issues? - about 5.
  • Who would like to contribute to BIM? - about 5.

General Comment:

Layers of interoperability - actually 4 - only 2 mentioned - each needs tools and development

  1. data - addressed by WP2
  2. models - addressed by WP1
  3. codes - not represented but had some work on Annex 60 on FMI that needs to continue
  4. services - grid, heat, etc. - get into business side, and things to provide with building - could be in WP3

Answer:

Codes is certainly part - could use FMI and mainly in WP3 - had hard time defining research topics to be defined around FMI for this community.

  • may not have enough momentum with only few people
  • these few people can be connection to make sure Project 1 is following standards
  • but want to keep using and building on FMI

Services - is co-simulation way to go? How scale up to test services aspects? Need to define core tasks and work on related aspects as far as possible to make impact.

Another comment:

All proposed is great, but need just a little more:

  • Collaborative design and engineering - construction part of industry ready to adopt - knowledge and competence in those companies - archs and engs drawings ignored and they create their own
  • at Stanford have virtual design and construction - collaborative design is essence of that
  • what we are doing is collaboration on energy design
  • Would expand package on testing - construction has knowledge and competence - need tools

Answer:

Two types of testing:

  • models fit real world
  • testing through design workflow
  • Project 1 to provide tools for performance mapping and data, but would be good to get more process to be sure tools used in right way
  • want to link to design workflow efforts, but don't want to try to expand to do everything

May want to add 4th task - what needs to be done in addition to this?

  • IFC model is most complete and only international standard that exists - but problems with it:
    • Missing some things, but main problem is problem of not knowing how to use it
    • Building Smart: forming task force to layout map to say what to do to IFC model to make it more useable

Answer:

Agree these are issues - but also need to stay focused - inlude use-cases and workflow within WP3 - but don't have resources to create roadmap.

Comment on CAD:

  • All tools are marketed and profit driven - no open-source - no interoperability
  • EnergyPlus geometry model came out of DOE2 - written two-page detailed with references proposal in changing archaeic geometry model in EnergyPlus
  • need to change this to streamline process of architectural design and simulation

Answer:

Could be part of WP3 on roadmaps for what's needed.

Question:

Moving WP1 to other energy carriers instead of just thermal - but also electrical?

  • no electrical engineers, lack of critical mass and resources, although various libraries for electrical systems exist in Modelica
  • most groups use other tools for electrical systems - is modelica best tool for this? - co-simulation will be important.

Answer:

  • Huge task to develop own electricity modeling
  • Scalability to large networks is still in question
  • Interfaces with other models will be important - cosimulation
  • Discussing Professor van Dreesen but not enough resources

Comment/Recommendation: Room is full of scientist who like solving complex problems but tools for industry. Makes sense to have technical advisory committee (TAC)?

  • can be aware and give recommendations and make early use.
  • if each subtask leader could form a TAC that will review internally - especially those who may be interested in using tool

Answer:

Good idea. Trane and MERL. - would be good to get people not directly involved to comment - could also get IBPSA Fellows

Additional comment:

How expose structure of model from manufacturers?

Get people to use tools before we're done - designers getting paid with time - not better buildings. Architects, Engineers who may be interested in using to give some focus.

Owners and Operators as well for MPC, etc.

See value, and understand, but have to need to do it to start to use.

FMI important so don't have to use it in Dymola or JModelica, etc.

Comment:

Take a step back - still difficult to work with IFC, standards, etc. - complexities of buildings is people from different fields not looking at building in the same way.

  • Energy modeling requires zone-based
  • HVAC is schematic
  • life cycle analysis - materials, etc.
  • communication between points of view
  • scaling from building to city brings in even more complexity
  • if want to have gateways to enable information tranfer - need to define information and gateway

Answer:

Could be discussed in WP2. Example from BIM:

  • BIM is 500 page document defining information for all trades and logistics
  • would be great to work on this.

Return to TAC:

  • how to organize? Where to meet? When to meet? How to get people involved? Who to meet?
  • perhaps once a year not with IBPSA review board
  • needs to be format attractive to people from all over the world

Answer:

From experience on TAC on Building Smart: Setup a charter for the committee - what are expectations and how do they operate?

  • look at whole organization or particular part - needs to be very specifically defined
  • do not need to meet face-to-face - virtually is ok with agenda and summary afterward
  • need to consist of people who are willing to talk to eachother - and who are invested in talking to eachother and project group
  • 5-10 people who are colleagues and share worldview
  • Vlado would be happy to provide names

Email ideas of names to Michael and Christoph to serve on TAC.

Workplan review, feedback, and commitment of work

Commitments

WP1

1.1 - about 5

1.2 - about 8 - need more?

Leaders should use mailing list (https://groups.google.com/forum/#!forum/ibpsa-project-1-announcements) to announce meetings of subtasks and use github for agendas/minutes/timing announcements

WP2

2.1 -

2.2 -

Email Christoph with name and project that can contribute to work-package

Questions

Question:

How collaborate between workpackages?

Answer:

People can overlap subteams and meetings - meetings among workpackage leaders

If models are broken in DESTESTS - knock on 1.1 door to say

Dissect work in projects - example: 4 pipe system model in library, but then also application work in dissemination task.

Also need to be sure to define applications early so that models can be developed, etc. Coming to common cases is hard - could be good to have brainstorming sessions among a few people.

Comment:

Meetings don't just have presentations or reports but also workshops to show on the PC how it works, how to use, training, etc. - small but practical sessions

Answer:

Not only for ourselves but also how to train people who will use the technologies.

Comment:

On website - could be good to have status indicators (red, green, yellow) and contact info for each project or parts of projects. Will help to have very open communication on github/repos - may also need to train people on workflow tools.

Comment:

Often in Annex 60: Model is asked for, but can't find resources for it. - need to be clear about who can do it.

Comment:

Could be helpful early on to define use-case scenarios so that big picture is defined and WP1 knows where to work towards.

Comment:

Visibility - what want to be known for? - goal is provide tools to be available for whole-building energy simulation and controls.

Comment:

MPC library - Picard had good approach to building MPC.

Comment:

New use of IBPSA library in EdF?

Dissemination Strategies

  • Having good training material - takes time and money to develop
  • Provide platform for posting all of these materials so that we can direct students and new users to - important for each subtask too - tutorials, video clips, taching materials
    • git, repo, structure, etc. - agree on the media
  • How build on generalized Modelica approaches? - leverage their materials.
  • Record what we do? - Week workshop in France - was it recorded? Could be and posted.
  • Still visibility issue - You have question on building simulation? - know that for some answers, we are the source.
  • Well organized TAC will go a long way to achieving goals.
  • At next meeting, need to have a definition for what mean by dissemination - including leader - email if something would like to do.

Next Meetings

Meeting every 6 months - Feb/Mar and Sep/Oct

Send out doodle poll to find a date.

Organize with local contraints and local IBPSA conferences.

Look for volunteer to host next meeting:

  • provide rooms - one large to hold 50-60 plus 5-6 breakout rooms for 5-10 people with projector
  • organize dinner
  • for cost, have donation from Sponsoring Participants (currently Mitsubishi MERL) to offset - typically for rooms - but at university typically free
  • James in Dublin, Ireland could host 2nd
  • EdF could host after 1st - south of paris
  • London
  • Aachen, Berlin
  • University of Colorado - once Wangda settled
  • Host in Flanders - KU Leuven

Next meeting - First half of March in Berlin

To convince funding agencies for travel - easier for Annex but could be tougher IBPSA.

  • agree on meetings earlier and really try to attach to other conferences to minimize travel, money, carbon, etc.

Next Steps:

You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.
Press h to open a hovercard with more details.