[Equest-users] EEM vs. new model

[Nick Caton]

Hi Matt!

If your primary goal is to compare the relative performance of two systems on the same building, then you can rest assured you've made fine decisions so far.  That you're simultaneously tasked with "validating" one choice over another might complicate things in presentation of your results - it depends on how the cookies crumble.  More later.

eQuest is a capable tool to this end, but I might have cautioned you at the outset that eQuest can quickly become unwieldy for "a simple comparison" when trying to compare certain system types, especially for newer/intermediate users.  That said, you are clearly well-invested into this model, and why would anyone want to miss a learning opportunity, right =)?

General rule:  A more representative model will always produce a 'better' result - the corollary is that 'better' is not always what you're hoping for.  "Degree-of-accuracy" is always a balancing act for any model - it's clear you're taking an active role in deciding when/how to improve the model's degree of detail, however.  That's enough to satisfy me.  So long as you personally feel the model is "representative enough," that's the key litmus test when you're the one presenting the results ;).

I would not suggest stopping and building a 90.1 baseline model provided with this extra context, but you could use 90.1 as a broader reference to ensure you are making a fair comparison.  For example, 90.1 Appendix G has a section discussing unmet hours and establishes a threshold for both the (90.1) baseline and proposed models.  Have you made a similar check between your chiller and DX models to ensure they're on similar footing?

I almost always build my models from DD wizards instead of SD so my usual workflow may produce different "normal fixes," but I commonly need to spend extra time reviewing the hydronic inputs in detailed mode to ensure any chiller system is being correctly modeled.  Give your chiller, pump, and loop inputs  a good scour, particularly pump operation/scheduling inputs.  You mentioned also changing the quantity of airside systems for the DX model - it might be worthwhile to review the schedules applicable to the consolidated system to ensure its operating as a single large AHU would.

When all is said and done if you still feel uncomfortable with your inputs, I've often found it productive to have the responsible mechanical designer scan through typical waterside/airside inputs alongside me to identify anything that appears off.  I have found pulling other designers in is a good QC practice for all skill levels where the modeler didn't do all the work him/herself.  I find it easiest to plan on missing some nuances along the way (Murphy's law), and allow those most intimate with the design to collaborate with me to pick any loose threads up.

So full-circle... regarding your 'intended' results:  Accept that the "right" answer can ultimately swing either way.  Whenever performing a comparative study, keep in mind all such results are built on unknowns/assumptions.  Where the results are close, it's not uncommon to conclude there are multiple right answers.

~Nick
[cid:489575314 at 22072009-0ABB]

NICK CATON, P.E.
SENIOR ENGINEER

Smith & Boucher Engineers
25501 west valley parkway, suite 200
olathe, ks 66061
direct 913.344.0036
fax 913.345.0617
www.smithboucher.com

From: Matt Clough [mailto:mclough at kme-inc.com]
Sent: Tuesday, April 24, 2012 7:37 AM
To: Nick Caton; equest-users at lists.onebuilding.org
Subject: RE: EEM vs. new model

I suppose I should have been more specific in my description:

By baseline I mean the proposed design.  So the program outputs 2 sim files - the base(proposed, CHW) case and the iterated (DX) case


We want to use eQuest to validate the system choice on this particular project.  This is work that was started before I joined, but I have been selected to try and use eQuest to compare the performance of a chilled water system and a DX system.   I'm not the most experienced energy modeler, but it seems like a task that would be a good application for eQuest.

My approach was to simulate the energy consumption of a building that is similar to the actual design in orientation, shape, and square footage.  I initially built the model in the SD wizard, using most of the default values (building envelope, activity usage, etc.), autosize the equipment, and use the EEM to compare the two systems.  The idea was that when the building construction and geometry are equal, we can look at the relative energy consumption of the two systems.  This initial simulation reported that the DX system would be more efficient, which is not what I had expected based on the design that had been selected.
So I figured that maybe my model was too generic, too simple.  I created zones based on the plans, and made the geometry more representative of the actual building.  These changes produced reports that suggest the chilled water system is more efficient than the DX.  I realized that the EEM was incorrectly modeling the change - it was treating the chilled water system as having all 8 air handlers, but was modeling the DX system as one AHU.  So I created a duplicate model that has a DX system (autosized and using equest defaults).
The comparison of these two models suggests the DX system is slightly more efficient.  Unfortunately, my lack of experience makes me doubt the validity of this comparison - especially considering that the engineers selected a chilled water system for this project.

So I suppose my questions are: Does my approach seem valid? Is this an appropriate use of the software?

After taking some time to write this note, and with a little rest, I wonder if comparing the proposed design to an ASHRAE 90.1 baseline would have been a better approach.

Regards,
Matt Clough

________________________________
From: Nick Caton [mailto:ncaton at smithboucher.com]
Sent: Monday, April 23, 2012 3:40 PM
To: Matt Clough; equest-users at lists.onebuilding.org<mailto:equest-users at lists.onebuilding.org>
Subject: RE: EEM vs. new model

By context, I gather by 'baseline' you mean something other than your proposed design, such as an 90.1 Appendix G baseline or similar.

If that's the case, my general advice is that you should much more heavily weight the comparative results using your "actual/proposed" model.  A baseline per above inherently diverges from the actual building materials, lighting power densities, solar load shading, and so forth.  It's not the same building.

Best to choose systems that most efficiently handle your actual project.  Don't get too hung up what works best for an imaginary building ;).

~Nick

[cid:489575314 at 22072009-0ABB]

NICK CATON, P.E.
SENIOR ENGINEER

Smith & Boucher Engineers
25501 west valley parkway, suite 200
olathe, ks 66061
direct 913.344.0036
fax 913.345.0617
www.smithboucher.com

From: equest-users-bounces at lists.onebuilding.org<mailto:equest-users-bounces at lists.onebuilding.org> [mailto:equest-users-bounces at lists.onebuilding.org] On Behalf Of Matt Clough
Sent: Monday, April 23, 2012 1:59 PM
To: equest-users at lists.onebuilding.org<mailto:equest-users at lists.onebuilding.org>
Subject: [Equest-users] EEM vs. new model

Salutations eQuesters,

I'm doing a simple comparison between a chilled water and DX system.  It would seem that this is a great opportunity to use the EEM, and the simulation predicted that the CHW configuration would be more energy efficient than the DX system.  To be conservative, I also made a copy of the baseline model and changed the system from chilled water to DX.  Using this method, the simulation predicts that the DX system would be more efficient (although not by much).  Which result is most accurate?  The building is a car dealership with a large showroom and an air-conditioned parts storage area.

The geometry and building envelope are not exact, but I made this decision because we are doing this for comparison purposes, so the geometry and construction will be the same.
I have let eQuest autosize both systems.

Sincerely,
Matt Clough
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/equest-users-onebuilding.org/attachments/20120424/428326aa/attachment-0002.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 1459 bytes
Desc: image001.jpg
URL: <http://lists.onebuilding.org/pipermail/equest-users-onebuilding.org/attachments/20120424/428326aa/attachment-0002.jpg>