request: thermal simulation

[teobo]

Hi,
maybe I understand this whole stuff of BIM/IFC not so well. Is it that representations of buildings, its constructions and above all connections are possibly yet usually nowadays available that detailled and over more exchangeable
that they meet a detail degree like:
https://www.dropbox.com/s/51fuh69dsfl2g ... -02-03.pdf ?
Yes?
Then one could maybe waste a thought on thermal simulation. Disregarding humidity, acoustics and statics one could convert freecad in tool of commercial scale in the thermal simulations field.
There is as well a free tool around that performs such analyses:
http://energy.concord.org/energy2d/thermal-bridge.html (us locales or default locale for the browser)
from http://en.wikipedia.org/wiki/Thermal_bridge
Unfortunately it is in java and I can not so easy feed it with freecad output. (Maybe someone knows how to trigger that.)
One could do whole building simulations as well as party simulations that by now can be done only by commercial tools.

Much ifs and whens.
What do you think of it?

[yorik]

Certainly interesting, but ithat kind of stuff represents a huge amount of work to implement...

[teobo]

Certainly interesting, but ithat kind of stuff represents a huge amount of work to implement...

It seems huge outcome.
I do not it agree this must be huge, it is a question of reaching the interfaces, providing a xml for each connection (given you got detailed connections) and maybe configuring the tool energy2d, which is not trivial, summing up the results each one, done! and one had a scripted prototype.
The current problem is to control the java tool from python. It has well - I really have no idea from java - some "service interfaces" that can control the application. reload configuration, run, stop simulation. Some script components are even missing: output results.
So in the end I do not see huge work in it.

Taking all the code and the visualization to freecad, yes, that could be - some work (;

[saso]

Personally I actually (at least in an long term view) also have quite a lot of interest in all types of such simulations and analyses (structural, thermal, light, sound, energy,...). Most of the time I see that suggestions for this type of tools, while for different systems (thermal, structural,...) are mostly for the simulations and analyses that are based on the FEM model. They are of course ok and specially visually nice, but they are generally also in the group of "complex" systems, that are generality hard to implement and use correctly.

There are however for all of this analyses also known "simple" methods that are accepted and used in the industries. This are generally all of the standard methods that are used in engineering calculations (structural, sound, thermal,...) that have been used in the industry for decades, before such things as FEM simulations were invented. The nice things about them is that they are generally more simple to implement and work with and at the same time give valid (accepted) results. While for more complex models they can also get complex, the actual biggest problem with them is that mostly if you want to get valid results from them they have to be implemented by standards that are generally not open and also different in different places (countries)...

But I still have hopes to see at least some of this in the future. First step would IMO be to build the analytical models, from the Arch elements. For the "simple" methods this are not FEM meshes of the complete designed objects, but simple lines and surfaces and as I know yorik already did a bit of this for the structural model

[teobo]

While for more complex models they can also get complex, the actual biggest problem with them is that mostly if you want to get valid results from them they have to be implemented by standards that are generally not open and also different in different places (countries)...

With the standards you are right, they are insultingly expensive. One needs to kiss up always to somebody with job at the university.
I find in general the situation is not that bad. It is like occ or coin, they are maintained by there own people. And I mean it is not the worst thing on earth when user's understanding of her software is profitable, imho.

[teobo]

Hi,
do not want to disrupt any discussion here, just leave some other two pence:

consider the lists of quotes below:

"IFC is an extension to the STEP format, and is becoming the standard in BIM workflows."
-Green Building XML (gbXML) is an emerging schema, a
-subset of the Building Information Modeling efforts
.gbXML is used as input in several energy simulation engines.[36]

It's wikipedia, and if you look up that list, you find not one free engine. Hey I might not believe in bim wash, but its not considered blasphemy believing in green wash. So the need for free cause open source modeler is there.
And where would the whole world come feed their ifc to energy engine?

Enough dreaming, get to sleep.

[saso]

Just a small note on the topic and to save the link for later

http://www.energy.gov/eere/buildings/ar ... ge-barrier
https://github.com/NREL/EnergyPlusRelea ... /tag/8.2.0
https://github.com/NREL/EnergyPlusRelease

[minimalexpert]

From the Energy Plus Documentation:

"EnergyPlus is not a user interface. It is intended to be the simulation engine around which a third-party interface can be wrapped. Inputs and outputs are simple ASCII text that is decipherable but best left to a GUI (graphical user interface). This approach allows interface designers to do what they do best—produce quality tools specifically targeted toward individ- ual markets and concerns. The availability of EnergyPlus frees up resources previously devoted to algorithm production and allows them to be redirected interface feature development in order to keep pace with the demands and expectations of building professionals."

This is great that the back end is largely done and FreeCAD has a great visual front end. I am a lay person though, so I don't even have a sense of what it would take to connect the two. I remain hopeful...

[paullee]

Seem interesting.

Anybody had tried what it can be done?

[PeteSmith]

Hello everybody.

OpenStudio is also a US Department of Energy backed software project. It appears quite modular, can import IFC, and has a Sketchup plug-in. It uses EnergyPlus and Radiance for various simulations. The IFC interoperabillity is through BimServer. I haven't used it on a project, yet, but I'm very intrigued.

http://nrel.github.io/OpenStudio-user-documentation/