[Bldg-sim] One model

[Michael Donn]

Hi All - again

When I created my long post on the Daylighting models I self-edited because it was looking like going on forever. For example, I stopped short of listing the full range of software that has been subject to validation exercises, and stuck to just those that have been validated against measured data. There is for example the validation suite created by the CIE to be an equivalent for lighting of the IEA BESTEST / ASHRAE 140 validation process for thermal simulation programs. To my knowledge, AGI 32 and the Integra software from Japan have undergone this validation process.

This most recent post returns our focus to energy simulation. I have over the past six months of a sabbatical at the Lawrence Lab in Berkeley CA come to know and appreciate the COMFEN program which is targeted at providing the power of the EnergyPlus program, very early in the design of a building.  Programs like EnergyPlus have inbuilt engines for calculating light distributions due to daylight. The COMFEN (http://windows.lbl.gov/software/comfen/2/index.html ) program from LBNL makes some of the daylight calculations within EnergyPlus accessible in these very early phases of the design process. This is fully integrated.

Clearly an add-on for daylighting to a thermal simulation program will not be the best lighting program. However, it should be sufficiently accurate for many purposes where thermal and daylight performance are combined.

Rob, you are correct about the su2rad/OpenStudio plugins for SketchUp. Because the OpenStudio plugin from DOE/NREL will read EnergyPlus input (idf) files, SketchUp is the potential integrator of COMFEN/EnergyPlus and Radiance. It works well here because of the elegance of the underlying template for the COMFEN EnergyPlus model. But even here it has limitations. The thermal simulation model representation of a wall is a set of 3D coordinates locating a surface in space and its orientation to the sun. The thickness and thermal properties are represented mathematically. Reading an EnergyPlus '.idf' file into SketchUp reveals windows with no frame and reveals with no depth - anathema to the good daylight modeler. If one creates buildings with depth for daylighting in SketchUp, one needs to specify thermal zones within the model before exporting to EnergyPlus.

The single BIM model that links to all analytical programs but is independent of them all is an ideal that has been worked on since the days of the COMBINE project in Europe 20 or so years ago. Many are working still on it. Logic suggests it is the most elegant solution to the dilemma of exchange of information between many different analytical programs. But this simple illustration suggests to me that the Building Information Model is not SketchUp. It is merely one on a long line of geometry creation tools. It is unusual in that it can be used to create lighting, acoustic or thermal models. But it merely creates a common starting point for these separate analytical tools.

M


Michael Donn
Director Centre for Building Performance Research
School of Architecture
Victoria University      +64 4 463 6221  work
PO Box 600              +64 21 611 280  mobile
Wellington                +64 4 463 6204  work fax
New Zealand             +64 21 611 594  mobilefax

From: bldg-sim-bounces at lists.onebuilding.org [mailto:bldg-sim-bounces at lists.onebuilding.org] On Behalf Of Rob Guglielmetti
Sent: Thursday, 12 March 2009 4:12 PM
To: Ross Harding
Cc: bldg-sim at lists.onebuilding.org
Subject: Re: [Bldg-sim] One model


On Mar 11, 2009, at 6:09 PM, Ross Harding wrote:


The ultimate goal for us would be to use one model for daylight and energy models, but I haven't found that quite that simple as most software requires varying drawing techniques.

Amen brother!  Another issue is that while the energy model has to include the total building -- all spaces and systems -- to be a complete picture of the energy use, a Radiance-based daylight model does not necessarily require all the spaces to be modeled simultaneously.  Indeed, the accuracy of the ambient calculation is directly affected by the maximum size of the scene, and so conducting a simultaneous daylight simulation of an entire building at an acceptably rigorous ambient resolution can be -- is generally -- time-prohibitive.

As mentioned in a recent thread on this list, Thomas Bleicher's "su2rad" plugin for SketchUp can export a SU model to Radiance format and there is also an E+ plugin to allow one to use the same SU model for E+ analysis; in theory, these three components (SketchUp, su2rad and the E+ plugin) allow for a "single model" approach to energy and daylight modeling. However, because the su2rad exporter wants to take the entire model and create a single Radiance scene description, that creates the aforementioned problem of the ambient calculation getting out of hand fairly quickly.  I suppose through intelligent layering and model structuring, one could create a model that could be exported to Radiance in "space components", but I'm not sure if this is compatible with the E+ plugin's layering/model organization requirements.

- Rob Guglielmetti
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20090313/452cc834/attachment-0001.htm>