[BLDG-SIM] User Friendly (was: Building Simulation in the U.S.)

Jason Glazer jglazer at gard.com
Wed Jun 13 05:57:43 PDT 2001


Lets not forget that the team of design engineers and 
architects working on the design of a building knows as 
much as anyone could know about how the building is 
supposed to work.  Through drawing and specifications, they 
have a created a model of how the components work and 
interact that can be turned into a working building in 
order to serve the needs of the occupants.  The complex and 
rich expression of how a building is supposed to be 
simulated is already present just in the wrong language.  
Allowing easy expression of this type of complexity in a 
building simulation program therefore is "simply" a user 
interface issue.  It is solvable.  My suggested general 
rule to follow: 

  The closer to the design team's way of expressing the  
  design the better.  

That said, the current state of the art in building energy 
simulation programs are not very close to the way engineers 
do express designs.  

For example, a schematic design with all flows between 
equipment and a sequence of operations for the sensors and 
controls is what is often used in practice.  Why isn't the 
building energy simulation software input more like that?  
In part, because in the past, the input requirements are 
driven by the algorithmic approach of the software.  This 
needs to be reversed to achieve "friendly" software - the 
algorithms need to be driven by the input requirements.  

Thermal blocks (a.k.a zoning) are another big issue, the 
level of abstraction needed to understand what a good 
thermal block plan would be for a simulation program 
requires an understanding of the simulation algorithm.  Why 
should a design engineer care?  Assuming that all surfaces 
are imported from a CAD program (a big assumption) and each 
zone is associated with a type of occupancy, why can't the 
simulation program find out the best way to break up the 
building into thermal blocks.  Someone could come up with 
an algorithm to do this.  

Another issue is libraries of materials and equipment.  It 
is in the best interest of the manufacturers of products to 
supply enough information for building energy simulation 
programs but they don't. What are the barriers here? 

a) Different types of information needed by each software 
program.  

b) Possible incorrect interpretation of data by the user or 
the software.  

c) Not a large enough market to bother with supporting 
those customers.  

d) Not enough specifying engineers asking for the data. 

The last point brings us back to why are simulation 
programs not used more.  One way to start solving these 
programs is to look at all the standards used to specify 
the different products and make sure the simulation 
programs accept the standard performance metrics as direct 
inputs.  DOE-2 cannot accept AFUE or EER.  Also product 
specifications should be examined for common methods of 
expressing performance and control of equipment and 
components.

What is left of the problem?  The unique domain of 
knowledge about the building simulation program itself.  
This is new knowledge that the engineer doesn't need 
normally to fulfill his responsibilities and thus it needs 
to be minimized.  Can it be?  I think so, the typical space 
loads and schedules for operation for occupation are easy 
lookup's in a library.  Engineers do that all the time when 
the specify building components.  Select a weather file - 
again easy.  The problem is all of these other inputs: 

ground reflectance, 
fraction of heat to space, 
weighting factor, 
infiltration coefficient, 
maximum glare, 
inside visible reflectance, 
convergence tolerances, etc..  

Any of these types of inputs require a engineer to learn 
new stuff.  I think most can be avoided completely, 
specified as part of a building component, or expressed 
differently in terms the engineer can appreciate.  

In summary, I assert that the almost all of the needed 
details of the building energy simulation are known by the 
design team of a building but they don't use building 
energy simulation programs partially due to the information 
being expressed differently.  The complexity that remains 
that is unique to building simulation can be minimized.

Jason
=========================================================
Jason Glazer, P.E.  mailto:jglazer at gard.com  847 698 5686
GARD Analytics - http://www.gard.com/
1028 Busse Highway, Park Ridge, IL 60068
Building Energy Simulation and Analysis
List Administrator for 90.1, GPC18 and BLDG-SIM

======================================================
You received this e-mail because you are subscribed 
to the BLDG-SIM at GARD.COM mailing list.  To unsubscribe 
from this mailing list send a blank message to 
BLDG-SIM-UNSUBSCRIBE at GARD.COM



More information about the Bldg-sim mailing list