Question-and-Answer Resource for the Building Energy Modeling Community
Get started with the Help page

What are your techniques for reducing unmet hours?

As many of you know programs such as LEED require total hours where the setpoint is not met to be below a certain threshold - e.g. 300 hours. Notwithstanding compliance, I find this to be a best practice for tuning a model. So what techniques do you find are most successful for reducing unmet hours (total, heating, and/or cooling) either in general or for a specific HVAC system? I'm specifically interested in the experiences of EnergyPlus/OpenStudio users.

edit retag close merge delete

I like what you did there!

( 2014-10-04 09:49:09 -0600 )edit

( 2014-10-06 09:33:11 -0600 )edit
3

I'm not sure if we are following similar rules to stack exchange, but this is a very vague question. All of the typical suspects: bad loads, setpoint tolerance, morning warm-up, etc, are below, but this also points to more difficult problems such as poor zoning an unintended/unpredicted heat transfer between internal zones. In general I think these forum styles are better served with specific questions and then your attempts to solve it. Open ended questions like this are better served on discussion boards.

( 2014-10-19 23:47:45 -0600 )edit

Do those 300 hours refer to occupied hours or total, ?

( 2021-05-27 22:47:55 -0600 )edit
1

@Brianna 90.1 doesn't explicitly state occupied or unoccupied, although I think the definition implies all hours. I personally reported occupied hours when I was doing LEED projects.

unmet load hour: an hour in which one or more zones is outside of the thermostat setpoint plus or minus one half of the temperature control throttling range. Any hour with one or more zones with an unmet cooling load or unmet heating load is defined as an unmet load hour.

( 2021-05-28 11:11:10 -0600 )edit

Sort by » oldest newest most voted

Troubleshooting unmet hours in OpenStudio

---UPDATE 7/21/2015---

This Measure automates many of the checks listed below and gives you a nice report summarizing the results.

Basic checks:

1. Look at hours unmet per zone table in eplusout.htm file
• Make sure that the zone with the problem is not a slave zone on a single-zone VAV system.
• Check your thermostat setpoint schedules for this zone to make sure values are reasonable
2. Check that design days are for the same location as the weather file
• Open the .osm in a text editor to look at this.
3. Check the tolerance for reporting unmet hours
• Try 2F and re-run. If tolerance is too tight, you can get false alarms.
4. Check design/sizing temperature for each plant loop.
• This is found by clicking the dashed line in the center of a plant loop.
5. Check the operational temperature in the plant loop's setpoint manager.
• It should match up with the design/sizing temperature for that loop.
• If you size the loop for 180F water but tell it to operate at 150F, the equipment won't be big enough during peak times.
6. Check design/sizing heating and cooling supply air temperatures for air loop.
• This is found by clicking the dashed line in the center of the air loop.
7. Check the operational temperature in the air loop's setpoint manager.
• It should match up with the design/sizing temperature for that loop.
• If you size the loop for 55F supply air but tell it to operate at 60F, the equipment might not be big enough during peak times.
8. Check that sizing design day thermostat is using a constant setpoint schedule with no setback.
• This is a 90.1 requirement. If your design day has a setback, your system will be oversized in an attempt to be able to go from setback to setup in a single timestep.
• Equipment that is significantly oversized might not operate properly at low load conditions.

More detailed checks:

Use the BCL Measure "Add Output Variable" to request timeseries data and inspect these data in ResultsViewer to see what's going on.

1. Examine the zone temperatures and their thermostat setpoints
• See how much the system is missing setpoint by
• A shoulder time when transitioning between setback and setup is common
• You might need to schedule your HVAC on a bit earlier to overcome this
2. Examine the unmet hours per zone
• See what times of day/year the unmet hours occur
3. Check sizing factor (safety factor) in model.
• This is found on the Simulation Setting tab

Other resources:

• IES VE has a very good guide that conceptually holds true for most simulation engines.
• Trane has a guide for Trace 700.
more

Hi Andrew, I was wondering if you can point me into the 90.1 requirement for the use of a constant setpoint schedule with no setback?

( 2014-10-16 17:05:12 -0600 )edit

"Try 2F and re-run. If tolerance is too tight, you can get false alarms."

I see in openstudio under output control reporting tolerances a unit of R. "Tolerance for time heating setpoint not met" example 1.0800 R.

Do you have any idea what the R stands for?

( 2017-10-19 14:13:00 -0600 )edit

degrees Rankine. 1 R = 1 delta F

( 2017-10-20 11:40:27 -0600 )edit

aparker, the link inside the "This measure" does not seem to work :/ {"detail":"Not Found"} How can I access this measure?

( 2022-06-22 04:26:39 -0600 )edit

EnergyPlus has a neat trick of being able to report out a ton of variables which can be viewed in the ESO file viewer. Usually if you are getting unmet load hours there is something wrong with your system as the previous answerers have pointed out. The easiest way to check your systems and what they are doing is to report out your node flow rates and node temperatures. For one hour of the day you can plot the temperatures and flow through your AHU. you can see the temperature and flow return from the zones, the outdoor air temperature and return, then the fan, the coils and finally what is being delivered to each space, it is here you can also see which spaces are reheating and which aren't (assuming you have a reheat system). You can do the same for your plant loops, it is clear when you do this, whether your flow is too high or non existent, whether your coils are working or not, whether they are meeting setpoint etc.., whether your outdoor air is coming in when it should be and stopping when it should be. Whether your fan is running constantly and so on.

By doing it this way you aren't relying on a check list and luck you are analysing what your system is actually doing and whether it is doing it correctly. More importantly you can point out to your clients any potential issues that they might have designed into the system that need to be discussed.

more

Regarding OpenStudio, I have found that the defaults for the Output Control Reporting Tolerances for heating and cooling are very low, 0.2K if you open an empty OpenStudio model. AHSRAE 90.1-2007 Addendum cr defines the reporting tolerance as half of the throttling range, or the room temperature difference between full heating and no heating or full cooling and no cooling. Of course this doesn't solve any modeling issues that may be causing hours to be unmet but it is important to understand how the unmet hours are determined and that they are being reported correctly.

more

( 2014-10-08 20:17:24 -0600 )edit
2

After finding the actual source, the correct reporting tolerance is half of the throttling range, or the room temperature difference between full heating and no heating or full cooling and no cooling. I made this change in my above answer.

( 2014-10-17 12:36:07 -0600 )edit

Fundamentally, unmet hours will be due to issues with either capacity or controls. From this basic understanding we can use various avenues to troubleshoot.

Capacity:

Here are some questions to ask when troubleshooting capacity issues:

• Are the loads in the spaces reasonable and what you expect?
• Are the capacities of the systems serving these spaces reasonable and what they should be?
• If the capacities are incorrect, revisit your assumptions or examine the autosizing to ensure it is working correctly

Controls:

If the capacities are as they should be then move on to controls:

• Are the heating/cooling sources actually operating?
• If not, when and under what circumstances are they off? Every day? Evenings only? Summer only?
• If they are operating, is the energy actually reaching or being removed from the space? (Is the damper open?)
• What devices/sequences/limits/etc could be causing this.
• If the devices are working correctly are the temperatures of the loops/fluids as they should be? What controls these temperatures in your model?

The particular methods used to answer these questions will depend on the software being used and may be a combination of standard or custom reporting.

more