Simulation Results

[Lisa Bramwell]

Dear Stewart,

One of my students used CONTAM to simulate the CO2, PM2.5 and HCHO concentrations within a building. The results for CO2 were comparable, with low NMSE but those for formaldehyde and PM2.5 had very high NMSE.  This is partly due to an inadequate description of human activity patterns by participants.  Also, his graphs had almost step-like responses to the onset of pollution episodes rather than soft smooth curves.   Any ideas as to where he may have gone wrong?  We are currently brainstorming.

Sincerely,

Lisa Bramwell.

[Dols, William Stuart (Fed)]

Most likely either the Output and/or the Calculation time step are too long to produce fine-grained results.

- Stuart

--
To unsubscribe from this group, send email to contam+un...@list.nist.gov
 
View this message at https://list.nist.gov/contam

[David Lorenzetti]

When comparing against field data, it can be important to include, in your analysis, the effective integration time associated with your measurement technology.  For example, if you measure formaldehyde using some kind of sorbent that you later analyze in the lab, then you aren't getting instantaneous snapshots of the airborne concentration (which is what CONTAM reports), but rather a smeared-out integral of the airborne concentration.

Even worse, both formaldehyde and PM have a lot of uncertainty in the source and loss rates.  For example, HCHO emission rate from indoor materials can vary with the temperature and humidity, and PM deposition and filtration rates depend on the exact particle size.  It doesn't surprise me that it is harder to tune the model to any measured concentrations, than it is for CO2.  On the other hand, you get more "knobs" in the model to tune with.

-Dave

[Lisa Bramwell]

Thank you.

We noticed that in the ach file the air exchange rates were pretty high of the order of 10^2, which is much too high.  I assume that the AHS is overcompensating and that knob should be turned down?

Could you send me a screenshot of an ach file with the air exchange rate less than 12 to share with our students?

Sincerely,

LB.

--

To unsubscribe from this group, send email to contam+un...@list.nist.gov
 
View this message at https://list.nist.gov/contam

To unsubscribe from this topic, visit https://groups.google.com/a/list.nist.gov/d/topic/contam/PXu9FiM3jDU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to contam+un...@list.nist.gov.

[David Lorenzetti]

> Could you send me a screenshot of an ach file with the air exchange rate less than 12 to share with our students?

I don't have an ACH file easily to hand, as my Windows machine is shut down right now.  But if the goal is to have the students understand CONTAM better, then a good exercise would be to have them construct a one-off model to check this out.

Model a single-zone building.  Give it two flow paths.  One flow path has a forced volume-flow element with a flow rate like 1 m3/h.  The other flow path is just a passive leak (e.g., an orifice model).  Run the model and inspect the ACH file.

Then, with that as the baseline, they can change model parameters such as the zone volume, the path multiplier on the path with the volume-flow element, and the path multiplier on the passive leak.  Predict and verify what happens to the air change rate, and the zone pressure, in response to these changes.

Then add to this basic model a contaminant, say with an interior source, and see how different choices of default concentration, initial zone concentration, and transport solver time steps affect the predictions.  This will start giving them ideas about the causes of your non-smooth concentration curves.

If they are just starting to learn CONTAM, these sorts of simple experiments can be invaluable.

-Dave

[Dols, William Stuart (Fed)]

I recommend that you and your students follow the detailed Introduction to CONTAM video tutorials and accompanying materials available on the NIST Multizone Modeling website.

This should go a long way to understanding how to utilize the program.

https://www.nist.gov/el/energy-and-environment-division-73200/nist-multizone-modeling/contam-video-tutorials

 

- Stuart

 

From: con...@list.nist.gov <con...@list.nist.gov> On Behalf Of Lisa Bramwell
Sent: Friday, May 10, 2024 1:57 PM
To: CONTAM <con...@list.nist.gov>