I’ve been running some tests between the annual loads of LBT1.2 and IES. I set up a simple office block in both models, provided similar simple input to both of them and observed the results. My expectation was that the results would match quite nicely, but it turns out they don’t. The difference for cooling needs is IES 109 kWh/m2; E+ 150.4 kWh/m2, a whooping 50%.
So right now I’m in the look for the culprit for this difference The first issue that I see is that mechanical ventilation keeps appearing even when set to 0. If you see below, this could very well be my 50% difference in cooling needs.
I made a very simple script based on the shoe_box sample, and I found out the same issue, see here:
shoe_box_annual_loads_issue.gh (101.6 KB)
If I would fix this issue, next step would be to do some sensitivity analysis comparison. I’m obviously not expecting a perfect match, but the results must be coherent. It would be an incredible step to get trustworthy monthly simulation results in 7 seconds, but I’m afraid it looks too good to be true
hi @pmcmm ,
Your model has zone mechanical ventilation due to the ZoneHvac:IdealLoadAIrSystem. When you choose the ideal load air system the object has a OutdoorAirEconomizer. The Ventilation Outdoor and the infiltration of you model are:
If you turn in the Outdoor Air Economizer Type to NoEconomizer
the result is:
But you will increase the cooling load, so this is not the difference between the two calculations.I think is more related to the EP Ideal System from and the system you have choosen for IES.
Economizer is something unheard of in Europe. It’s very confusing to set ventilation to zero and still have flow through an economizer. Thanks a lot for the input, that was it! Actually it decreases the cooling load as my building is located in Vietnam, where the outdoor temperature is always above 25 C.
The difference now between E+ and IES is 10%. That’s a very acceptable figure!
The US economizer is the Europe free cooling section of the air handling unit.
The default economizer is “DifferentialDryBulb” so if your model is in a wet hot climate sounds good the reduction.
EP let you specify the ventilation in simple way with the ZoneVentilation:DesignFlowRate object but this not limit the use of outdoor air to reduce the cooling load with the economizer.
Could you also show me how you got to graph the mechnical ventilation in my model?
The image below:
I still get an incredible difference in the mechanical ventilation, which is 25 kWh/m2 in IES and 50 kWh/m2 in E+… I suspect there is a problem in the conversion of the input in flow_per_person to actual ventilation flow rate.
Hi again @pmcmm,
I ran your shoe-Box in the HB_to_OSM component and I added the string with the two output variables.
Wow, I have never seen an economizer increase the cooling load like that. I guess that I have not done enough simulation in the tropics.
The humidity must be the culprit. So, even though the outdoor air temperature sometimes drops below the room setpoint, the outdoor air is so humid that you wind up increasing the latent load more than you decrease the sensible load. As @BestiaParda says, anything other than “DifferentialDryBulb” will avoid this case. I might recommend using the “DifferentialEnthalpy” option instead of “NoEconomizer” so that you can still get the free cooling when BOTH the air temperature and the humidity are below the room setpoint.
I progressed further in the comparison between a detailed IES simulation and the speedy HB Annual Loads. Still impressed with the results. See below, they are quite comparable.
However there is still an issue with mechanical ventilation I am struggling to understand, I hope you guys can enlighten me. I removed completely the mechanical ventilation following the information in this thread, but I found out that if I change the minimum cooling temperature in the HB ideal air component from 20 to 10 C, the internal gains coming from people doubles. See below:
I also found a bug worth noting. The HB Annual Loads is incoherent in the area used to calculate the values in kWh/m2 if an unconditioned space is used. See below, the results from balance use the area without the conditioned space, and the total_load and others use the total area, including the non-conditioned space.
Here is the file if you wanna reproduce the issues I am describing here.shoe_box_annual_loads_issue.gh (143.7 KB)
The different results between the two calculations are due to the infiltration rate not to the internal gains coming from people.
When the minimun supply air is 10ºC the HVAC system operates less hours to reach the thermostat temperature and you have more infiltrations (when HVAC system does not operate you have more infiltrations).
You can check the infiltration flow rate in both models and you can see that are different.
Hi Bestia, thanks for following up. I don’t see the behaviour you describe happening in my model. There is a variation, yes, but it’s from 0.023 to 0.025, not such a big difference. I guess it is related to the wind speed. See below:
I think I found the problem, it’s related to the dehumification. For some reason, even when I leave the dehumidification setpoint to 100%, the simulation is still controlling and keeping at 50%. I guess taht is being provided by the AHU hence, the additional 60% in Mechanical Ventilation.
In case you are curious, this is the result of the latest comparison:
edit: corrected cooling value from IES
Thanks for reporting the inconsistency with the floor area @pmcmm . I’ll change it to use the total floor area in the load balance shortly and post back here once it’s fixed in the development version of the plugin.
I just merged a fix to make the floor areas consistent between the balance and the heating/cooling load:
Now, we always use the total floor area (both conditioned and unconditioned).
You should be able to get it with the LB Versioner in an hour or so and it will be in the next stable release.
@pmcmm , When you change the minimum supply air from 20C to 10C, the cooling coil has to work harder to cool the fresh air to a lower temperature, and more water vapour will be shredded from the air. As a result, you see a higher Mechanical Ventilation load, and the RH is “controlled”.