Sub-metering in Option D

I think I was the one who also asked Marcus regarding this issue. I think the reviewer will look for the method of how we can reach the end-use energy data. Some of them might require only runtime record if the equipment consume constant power. I used to create IPMVP plan this way and try to find the solution for obtain the data for each end-use (fan, chiller, pump, light, etc.). Yes, the submetering is not mandatory if you can find to solution to extract the end-use from the whole.

Good to see someone reading the IPMVP! I am not sure too many folks submitting M&V Plans have done so.

Let me see if I can explain the the rationale for needing energy end use data. First of all I do not believe I ever said that projects had to install submeters. There are several ways to get energy end use data and submetering is just one of them. Sometimes it is not even actual energy use data as it could be data you would use to calculate energy use. You are correct that submeters or any direct measurements of energy use by end use are not required.

You will still need to calibrate your energy model by energy end use and have a data gathering plan in place to enable it. If you just calibrated to the monthly utility data by fuel type you could very easily be over predicting one end use and under predicting another and on balance the result may "appear" to match but it doesn't. You will find that it is virtually impossible to get each month to match within a reasonable range. When it does not match then you will need to have some kind of energy end use data to figure out the discrepancy. So the M&V Plan must include a methodology to gather the data needed by energy end use. It cannot be an afterthought that you will figure out after the results do not match. I can guarantee that the monthly data will not match in all cases so you are going to need some additional data. I think anyone who has calibrated an energy model would attest to this fact.

Recently we gathered some whole building data for a project we modeled that was an all electric building. Our model was within 6 kWh in the annual total energy consumption. Pretty good huh? Well it turns out we over predicted heating and under predicted cooling and on balance we got lucky. Even if each month was within range we would not have a calibrated model.

Unfortunately the language in the IPMVP is often a bit vague. It sets out broad goals and lets you figure out how to get there. Based on first hand experience I can tell you that you can't calibrate to the utility bills without calibrating to energy end use. Another way to think about it is to focus on the outcome. The energy model basically needs to be fully reflective of the way the actual building operates in order to be accurately calibrated. Without any energy end use data I would suggest that it would be impossible to do so.

Now the question becomes do we need to measure or meter each energy end use? Probably not. Very small ones that do not have a big impact on the overall energy use can be estimated. Maybe HVAC can be lumped together instead of trying to break out fans, pumps, heating and cooling. Some end uses could be derived from subtraction. This starts to get very project specific and how you do it depends on many factors.

If you can make the case in your M&V Plan that you can accurately calibrate the energy model without any energy end use data I would like to read how you will do it. If you think it through you will find that it is not possible unless you have clairvoyant energy modelers!

Yes you are correct. Here are some others.

You can take spot measurements and use that data to calculate the energy use. For example, if I know that the lighting is a constant load (i.e. always on) and the schedule is consistent (on at the same time, off at the same time) I could use a clamp on meter, measure the wattage of each fixture type, count the fixtures and do the math.

I mentioned subtraction before. For example, if you meter lighting, hot water and HVAC you could derive plug loads by subtracting the submetered data from the utility meter data.

Some information can be obtained through interviews with operational staff. Things like occupancy and schedules are often obtained this way.

Commissioning activities often yield useful information what can be used in calibration.

Hand calculations could be used to verify very small constant loads if necessary.

And do not forget to calibrate the weather file to reflect the actual weather during the M&V period.