Skip to content

2020 01 13 Task2.2 Coordination Meeting

e.fichter edited this page Jan 15, 2020 · 1 revision

When

January 13 2020, 5:00pm CET/MEZ (= 17:00 German time = 8:00am Pacific time)

Participants

  • Jérôme Frisch - interim work package leader (RWTH Aachen, E3D)
  • Christian Waluga (liNear GmbH)
  • Jes Stershic (University of Colorado)
  • Karl-Heinz Häfele (KIT)
  • Eric Fichter (RWTH Aachen, E3D)
  • David Jansen (RWTH Aachen, EBC)

Agenda

  • Performed work in the last months
  • Next online meeting
  • Topics of Breakout groups in Canada
  • Participants in Canada

Discussion and Minutes
Today's discussion follows up to the online meeting in October.

Performed work in the last months

  • Jérôme welcomes the participants
  • There are too few participants to distribute responsibilities today, but discussions are ongoing
  • Eric and Christian worked on an Octree algorithm. First, the IFC file is opened and filtered by classes. The brep geometry is then built up based on the semantical IFC data and later converted into a tesselated triangulated form. These geometries are imported into the octree, that finds spaces and a selection of potential bounding products. Eventually, this information shall be used to extract first level space boudaries on next steps.
  • Karl-Heinz presents an implementation of an MVDXML in the FZK-Viewer to filter the IFC file by different semantical criteria. This is part of the modular workflow as defined in the last meetings.
  • Karl-Heinz shows presentation summarizing the workflow from IFC to IFC with space boundaries, as discussed in the last meetings. It was agreed to use only full non-split IFC models
  • Discussion on which data structure to use to represent the building's geometry. Curved faces have to be tesselated. Planar faces could be hold in its’ original polygonal form or in a triangulated form. Advantages of both approaches discussed. For example, triangles could be used for fast and simple geometrical tests, while trianglulation raises the number of faces to be checked.
  • David reports about the BPS part of the BIM2SIM project. A tool/workflow is developed which creates multizone models with TEASER by using the information from the IFC file. The approach is able to use the common enrichment methods from TEASER, but first it checks for existing data in the IFC. The user will receive an ready to run modelica model. Current steps: Integration of all possible information stored in IFC depending on the used export tool (Revit, Archicad). Semantic detection of thermalzones/spaces in IFC

Topics of Breakout groups in Canada

  • First group session can be more general about current states of research
  • The next sessions should into more technical details, e.g. coding techniques, algorithms and technical aspects
  • KIT will prepare material, that could be discussed by the participants
  • Eric and Christian will present current approaches for geometric algorithms
  • A video conference in the afternoon hours would help to expand discussions to a broader audience, that can't attend the meeting in Montreal

Participants in Canada

  • Jérôme and Eric: participate
  • David and Christian: pending decision
  • Karl-Heinz: will not participate

Next online meeting

  • Next online meeting will take place in mid of march
  • Please take part in poll until 31st January
  • Link to poll
  • Agenda will include plans for Canada and current implementation states
Clone this wiki locally