DOE commercial building - some issues with Openstudio measure

asked 2020-05-20 11:09:29 -0500

Mehrdad Vojdani
1125 ●4 ●12 https://www.linkedin.c...

updated 2020-05-23 16:33:50 -0500

Dear Developers and Engineers,

Our research group is going to perform an energy performance enhancement study on existing buildings. We found DOE commercial buildings suitable for our case study since it provides so many comparison opportunities.

Our study does not concern with HVAC equipment and control but we need it to be and work based on the latest version of ASHRAE 90.1. We decided to go with OpenStudio measures which simulate the mechanical (HVAC) system as well as constructions. I don't know how these measures work and whether is there any validation process or not. But according to this video, the measure named "Create DOE Prototype Building" was chosen.

As we were checking the simulation results, we found out some issues that I need to share with you (LargeOffice-ASHRAE 169-2006-2A):

A high amount of unmet hours was observed. I believe it can be due to low temperature tolerance. According to Matt Larson's answer to this question I changed the temperature tolerance to 0.556C. But it didn't solve the problem. here are some pictures of the Openstudio results and here is the results of unmet hours troubleshooting. According to troubleshooting, a temperature difference of 2.5 F would reduce unmet hours significantly. Isn't it more of a reporting problem than an operational problem? Any idea to solve it?
OpenStudio didn't provide any warnings concerned with the HVAC sizing or simulation. But Energyplus err file contained plenty of warnings that worried us. here is the IDF file and err. To be specific, warning associated with plant sizing and simulation seems pretty serious. Since we are not HVAC engineers, we would appreciate any advice to ignore or fix these issues.

Regards.

PS:

I believe I found some problems with the measure. In the IDF file, I found that CLGSETP_SCH_NO_SETBACK Summer Design Day with some setbacks during unoccupied hours is used for sizing calculation. While CLGSETP_SCH_NO_SETBACK Default and CLGSETP_SCH_NO_SETBACK Default|WntrDsn|Sat|Wkdy Day with no set back is used for simulation. I also checked with IDF files provide here and I think I'm right about it.

edit retag flag offensive close merge delete

add a comment

4

answered 2020-05-20 14:47:36 -0500

mdahlhausen

9314 ●17 ●11 http://openstudio.net/

updated 2020-05-21 08:26:58 -0500

Aaron Boranian

13830 ●76 ●24 http://bigladdersoftware.com/

For unmet hours:

The eplus html and unmet load hours troubleshooting still show you are using a 0.36F tolerance. Change this to 2F. This will get rid of most unmet hours.
occupied unmet hours are quite small; there is one zone with > 300 occupied cooling unmet hours, and this is because there is a spikey cooling setback and some residual system osscillation during unoccupied periods from the system during on/off. Either set the cooling setpoint to constant, or make the setback less extreme and for a longer duration.
check that the default schedule doesn't show residual occupancy (~5% occupied) outside of the HVAC operation schedule. EnergyPlus will report these as occupied unmet hours, even if the building is largely unoccupied.

On the warnings:

There is an oscillation issue in your cooling system and condenser loop, which is likely a result of night cycling and the rigid tolerance threshold above. The system is cycling on/off quickly becasue of the low tolerance at a rate smaller than the simulation timestep.
There are no severe warnings, just issues where design and user sizing values do not match on your VAV terminals. Check that the sizing values are aligned.

For your research project, ASHRAE 90.1-2019 is not yet available as a template option for Create DOE Prototype Building. You will need to update HVAC efficiency, envelope, lighting, and other properties in your model to match 90.1-2019.

edit flag offensive delete link

Comments

Sorry @mdahlhausen, the troubleshooting results were obtained with 1F tolerance (idf was a little bit older). The results with 2F tolerance seem ok. I found some problems with the cooling setpoint schedule of the measure. I added some details to my question. Please consider them and let me know your opinion about it. According to this the occupancy issue exists. Changing them to zero seems ok for me.

Mehrdad Vojdani ( 2020-05-20 19:27:11 -0500 )edit

1

@mehrdad_xyz try setting the occupancy from 5% to 0 and see what happens.

mdahlhausen ( 2020-05-20 19:41:49 -0500 )edit

On the warnings: by the temperature tolerance do you mean OutputControl:ReportingTolerances or the input in the Buiding object? the first one can be used for unmet hours. but the second one I think effects the condenser operation. Also If you search 'Maximum iteration' there 30 times that it reached the maximum limit. Is it normal?

Mehrdad Vojdani ( 2020-05-20 19:47:26 -0500 )edit

@mdahlhausen, your points about unmet hours were quite effective! to be specific, for the above-mentioned case study, check this report to see the improvements.

Mehrdad Vojdani ( 2020-05-20 23:28:29 -0500 )edit

add a comment

3

answered 2020-05-20 21:33:42 -0500

Jian Zhang
456 ●1 ●3

updated 2020-05-21 08:29:07 -0500

Aaron Boranian

13830 ●76 ●24 http://bigladdersoftware.com/

Hi there, as a member of the OpenStudio Standards Gem development team, including the "Create DOE Prototype Building" measure, we have not gotten to the level of model review to reduce unmet load hours down to ~300 hours/year as some modeling guidelines suggest. Here is a good and general check list for you to consider. https://unmethours.com/question/359/w...

In addition, here are a few known or possible causes for the unmet load hours in the prototypes:

The night cycling setting may be set to cycle on for certain period of time. If it doesn’t sync with the of the ending of a time step, during the remaining time the HVAC is off, which can cause unmet load.
For VAV systems using a fixed minimum box minimum, the heating load may not be met because the box is sized for cooling.
When supply air temperature (SAT) reset for an VAV system is used but the fan sizing is based on regular design SAT, there could be unmet load.
The infiltration schedules (mimicking the door infiltration) for design days should to be checked for sizing period.
There are 90.1 requirements to temporarily disable HVAC or setback thermostat when the someone opens the balcony door during favorable ambient conditions, interlock the HVAC with door switch. The modeling strategy to capture the savings may have been implemented in the OpenStudio prototypes. During the first hour when the thermostat is restored, the load may not be met.

There could be a number of reasons for unmet load hours. Some of them are unavoidable to capture some modeling strategies. Our development team have a list of things to review and unmet load is one of them. We will continue upgrade the measure.

edit flag offensive delete link

Comments

Thanks a lot, @Jian Zhang.

Mehrdad Vojdani ( 2020-05-20 23:30:53 -0500 )edit

add a comment

DOE commercial building - some issues with Openstudio measure

2 Answers

Comments

Comments

Your Answer

Careers

Question Tools

Stats

DOE commercial building - some issues with Openstudio measure edit

2 Answers

Comments

Comments

Your Answer

Careers

Question Tools

Stats

DOE commercial building - some issues with Openstudio measure