Hi everyone,
I am looking for an energy simulation programm that calculate heating and cooling loads, moisture transport and where natural ventilation can be activated. The software should (if possible) also have a bank for structural steel componenst such as steel joist, decking... or other structural component that are used in high-rise construction.
I use to work with DEROB-LTH (swedish software), but it does not take into account moisture transfer, and I am looking for something more efficient and developped. I was thinking of TRNSYS, Energy plus or ESP-R. Does anyone have other suggestions ordoes anyone has used these software and could tell me if they respond to my criteria?
Thank you
Madeline Leroux, M.Arch
Madeline,
a lot of software might claim to do all these things. But between claiming
and actually providing good estimations is a world of difference. The main
issue is that computational power is not at the level where a building cna
be subdivided in a trillion nodes that can each be calculated
simultaneously. What is the current general approach is that a zone or a
building material is considered as a single node, sometimes the zonal air
is subdivided in multiple nodes.
To give a good estimation regarding your moisture transport I assume it is
crucial to have an accurate calculation of zonal surface temperature. And
that is extremely difficult: transmission in the building structure can
be wel modelled, radiation could (at least theoretically, but of course
things change as soon as the room is furnitured, decorated and people are
using it) and convection is an even bigger problem.
I would start with some recent projects on heat air and moisture transport,
where ESP-r had been coupled to Fluent (If I remember well). Check for IEA
ECBCS annex 41. I am sure they have done some analysis of available
software and based their selection on potential to perform the task at hand.
Good Luck!
Leen
Energy Plus has the most power and flexibility, as well as great support, a
combined heat and moisture algorithm and allows for the creation of your
own custom libraries. You might get a free month of design builder to get
the hang of it..
Best
*Jeremiah D. Crossett*
The answer to your question depends a bit on what you mean by moisture transport:
If you mean moisture carried by the air as it moves from one zone to another and from outside to inside
via infiltration and ventilation ESP-r does this for scheduled zone air flows and for air flows calculated from the
solution of a flow network. A flow network can also be used to track contaminates (if additional information is provided).
And you can impose lots of different kinds of control on components within a flow network to approximate
user or automated controls for windows in natural ventilation schemes or hybrid ventilations schemes.
And for utter geeks there is the in-built transient (it adapts wall functions at each time-step for best fit
to changing flow patterns) 3D CFD solver that can have transient moisture or contaminate sources within
it and which can update its boundary conditions at each time-step.
If you mean moisture transport into and out of constructions within a building, ESP-r will also calculate this
and in validation tests has been seen to be quite close to that predicted by that well advertised application
from the good folk in Germany. That facility is HEAVY on the geek side. AND the limitation for most people
is that there is very little in the way of measured data about the dynamic response to moisture events for
materials. Without that information even the most clever solver remains unused.
It is quite possible to add in you own preferences for high-rise construction types if you do not find what
you are looking for in the basic distributed databases. And you can be as explicit as you want - e.g. if
you want to treat ceiling or floor voids explicitly so that you can implement structural cooling schemes
in your zoning as well as constructions and in the flow networks and controls and heat transfer
coefficients.
You did not state the complexity of the models that you hope to work with. ESP-r does some kinds
of complexity well. Users tend to avoid 'huge models' until they have a strong need and
mature skills.
We view ESP-r as a tool focused on the needs of the research community. Open and documented
and continually evolving. Users with strong opinions about the physics can coerce it to do
stuff that is quite eyebrow raising. Probably two score of PhD thesis have been based on it. Runs on just
about any computer (not yet on Android). If what you are trying to do has a research flavor then it has much to
recommend it. If you just want to apply it in consulting it might be able to handle mixes of design
ideas that commercial tools do not yet support.
I used to work with and code DEROB and I view ESP-r as generally more capable and covering more
analysis domains. Certainly the development community is larger. Its been used in validation exercises
for longer than most of us can remember.
Regards, Jon Hand
+++ Jon Hand [2012-05-17 17:48 +0100]:
I tried to package ESP-r for Debian/Ubuntu a couple of years back, but
after some hours of struggle I hadn't even managed to build most of it
myself, never mind get anywhere close to a useful packaging.
The makefiles seemed to have serious bitrot, and I got the strong
impression no-one outside of Strathclyde had built it for a long time.
I gave up and found something easier to do :-)
Assuming I was inspired to try again, because it does seem like very
useful software, and your comments suggest that it is still actively
developed, would you be a good person to talk to about it?
Wookey