Skip to content

2018 07 03 Task 3 Coordination Meeting

Jump to bottom Edit New page
InaDJ edited this page Jul 30, 2018 · 15 revisions

Task 3 Coordination Meeting

Date: July 3, 2018, 5pm Brussels time (UTC+2)

Agenda

  • Progress of Building modelling subgroup
  • Progress of Network modelling subgroup
  • Discuss links and needed steps
  • Start preparing Paris meeting

Meeting information

Minutes

Present

Present: Felix Bünning, Ina De Jaeger, Ilaria Vigna, Alessandro Maccarini, Dirk Saelens, Bram van der Heijde (notes), Konstantin Filonenko

Excused: Michael Mans, Peter Remmen, Artem Sortnikov, Christoph Stettler, Annelies Vandermeulen

Building modelling subgroup

  • Had a good meeting
  • Alessandro found some differences between his model and Ina's model, although he ran some BESTEST cases for both Buildings library and IDEAS library and these are in good agreement, he contacted Michael W and Filip J to check for solutions (buildings vs. IDEAS)
    • Seems to be problem with window models
    • South orientation seems fine, but east and west orientations might cause the discrepancies? Incoming radiation is the same
    • Check if solar radiation can also exit the building at the opposite side? For low azimuth (east and west)
  • Next steps?
    • Replace nodes one by one by a different heat demand profiles
    • Different building typologies (apartment, office, combination of single family dwellings)
    • Building description will be valid for TEASER as well as human-readable
    • Peter and Michael will contribute with AixLib/TEASER models
  • Discussion
    • How to make sure that there is enough variation?
      • Vary building age, occupants, building typology
    • Typically tertiary buildings have very sophisticated systems. How do these interact with the thermal grid? (instead of ideal heating)
      • Keep simple for now (constant efficiency,...)
    • Research plan/Roadmap?
      • on longer term
      • define goals for ourselves by Paris meeting, defined per subgroup
    • What accuracy do we want?
      • More than two zones? Non-ideal heating?
      • How to determine accuracy? > Compare multi-zone models to 2-zone models to 1-zone models

Network modelling subgroup

  • Bram presented progress of group
    • Subgroup meeting (see below): assumptions, aim and KPIs
    • Consumer model
    • Different approaches for network by Felix and Konstantin
  • Felix will adapt automatic dimensioning script considering different types of demand
  • Michael is going to implement uesmodels automatization
  • Validation?
    • Konstantin needs better structured data in order to start validating
    • Stochastic algorithm instead for heat pump model
    • Discuss with Mathieu Schumann during Paris meeting?

Actions

  • Buildings: check discrepancies + define other typologies
    • Everyone proposes definition of certain type (will be discussed in next subgroup meeting), then others model this proposal + compare + send demand profile of this type to network group
  • Network: Compare different approaches numerically
  • Need another meeting before Paris meeting
    • Think about free presentations
  • August: subgroup meetings
  • September: coordination meeting

Doodles for next meetings

Network modelling subgroup meeting

Agenda

  • Boundary conditions and assumptions
  • Key Performance Indicators
  • Division of work

Meeting information

Minutes

IBPSA WP3: Network modelling Meeting minutes – 5/06/2018

Present

Konstantin Filonenco, Felix Bünning, Michael Mans, Annelies Vandermeulen (notes), Bram van der Heijde

Boundary conditions and assumptions in network model

  • Pumps
    • Decentralized pumps at first
    • Valves and centralized pumps added in a later stage
    • Bypass valves: functionality can be added to control of decentralized pumps, so they’ll be left out for now
  • Control
    • Constant temperature difference across substation (20 K)
  • Network temperatures
    • One case for now: 50°C supply temperature
  • Pressure drops
    • Start with zero
  • Ambient temperature
    • Used to model pipe heat losses
    • Pipes are underground, so constant temperature is a good assumption (12°C)
    • Add ground model later on (Kusuda)
  • Heat generation
    • Ideal heat source

Goal of the network modelling

  • Comparing different control strategies
    • Using different pipe models? No, presently everyone is planning to use IBPSA models
    • Creating basic models on which we all agree to include in IBPSA library, different models can be added later on

KPI’s

  • Heat injection
  • Heat losses
  • (Pumping power)
  • (Total water mass passing through the heat generation)
  • (Network temperatures)
  • (Substation performance)

Next steps

  • Making network model
  • Split in two parts:
    • Network side: Felix, Michael, (Konstantin)
    • Consumer/control side: Bram, Annelies
  • Coordination through Github, no meeting planned until coordination meeting unless something interesting pops up

Varia

  • Use Marcus Fuch’s open-source code
  • Contribution to WP1: development of substation models?

Building modelling subgroup meeting

Agenda

  • Different modelling approaches for the building models
  • Next steps in the building model
  • Division of work

Meeting information

Minutes

IBPSA WP3: Building modelling Meeting minutes – 11/06/2018

Present

Alessandro Maccarini, Konstantin Filonenco (notes), Michael Mans, Ilaria Vigna, Ina De Jaeger

Progress on the tasks

Ina and Alessandro manually created a two-zone single family dwelling model in IDEAS and Buildings, but they do not agree as good as BESTEST models. Konstantin started modeling in AixLib.

Activities before the coordination meeting:

  1. Side activities to cover previous goals: Michael will help Konstantin with AixLib. Ina: we want to see what are the differences between the models, Alessandro: we still have discrepancies with Buildings compared to IDEAS. Therefore, Ina and Alessandro modelled a simpler version of the house to see where these deviations come from. He will make an even simpler model both in Buildings and IDEAS himself as a final check.
  2. New goals: new models both on component (buildings) and system (network topology) levels.

Discussions

  1. Buildings (in general, which types of buildings to include)

    Alessandro: (a) Different typologies are important, (b) Stick to 1 or 2 zone to use with existing TEASER code, (c) No air exchange between the zones, just heat transfer

    Michael: (a) More zones and gains are too early, (b) Made first steps in multiple thermal zones coupling in TEASER, but still on the way, (c) We can integrate and investigate 1 zone model. It is fine for DESTEST, (d) Good to compare TEASER approaches: use branches with Ina’s code to contribute

    Ina: (a) Can be challenging to include office before we finish with residential, (b) 2 zone are good for now, because we have TEASER code for it, (c) by using 1 zone, we would possibly miss important details, should not make so simple, (d) Maybe share the code to contribute to official TEASER version with the 2 zone model after discussing with supervisors

  2. Archetypes

    (a) other single-family dwellings of other types: semi-detached dwelling, terraced dwelling (node aggregation to combine loads for fx four dwelling), Ina

    (b) Case studies TEASER with AixLib are rather advanced: Office - 7 zones. Institute lab zone with an air handling unit - simulation becomes more complex (motivation: investigation of the research center near Aachen - study on area friction, archetypes), Michael

    (c) TEASER has a development branch with TABULA integrated into TEASER. When we agree on topology, we will have opportunity to integrate any TABULA data in TEASER. Belgium archetypes (Ina: 80s, rather high thermal heating load building) can be a testing material, then probably better go to Germany (as this is standard in TEASER) and more modern case studies (4th generation thermal networks). Michael

    (d) How is TABULA implemented in TEASER? TABULA can go slightly beyond just U-values: Recalculation of layer thickness based on U values and layer material (not all of it is available in TABULA, therefore, some assumptions are made: brick thickness which is always the same). Certain steps are automated, others must be supervised for TABULA/TEASER, Michael

    (f) Ilaria: apartment blocks

  3. Topology: Coupling with the district network model

    a) We should firmly establish topology (also with network group), to use for different archetypes, Michael b) Mix with other archetypes: for now only one building -> replace one node by one office building for instance, Ina c) Step by step replace thermal loads in each of the 16 fixed nodes, Ina d) Make possible transformation to all used libraries via TEASER, Ina e) Aggregate nodes of different dwellings into 1 (for now), so we keep the 16 nodes, but only change gradually the heat load

Actions

  1. Alessandro models an office building (TEASER example?)
  2. Ina: integrate other typologies of single-family dwellings in automated TEASER process (ensembles of different buildings we can choose from, including terraced, semi-detached and detached buildings)
  3. Ilaria proceeds with apartment block (now apartment with 2 zones)
  4. Konstantin proceeds with AixLib model of a single zone and manual network model (Michael)
  5. Alessandro is going to visit Aachen to make a TEASER implementation of Danish TABULA together with Peter. Michael invited others as well =)
  6. Alessandro will send the climate file in MOS-format and Ina will commit it to github

Deadline

Discuss with whole group on Coordination meeting when to plan the next subgroup meeting. We like the current setup (of network subgroup and building subgroup) very much.

Add a custom footer

Wiki

Main project site

https://ibpsa.github.io/project1

Clone this wiki locally