Simulating with real sensor data in OpenStudio

edit

Occupancy schedule is very easy to change into a model input. Dealing with outside air temperatures is more challenging. If you start changing temperature values without other values you can create a lot of problems. You may just want to get an AMY file near your building for the correct year and validate that it is reasonably close to your model. If you have ventilation or maybe even infiltration their might be a way to use EMS to override the inlet temperature but that won't change the temperature the envelope is exposed to. If you do edit weather file I would use one of a number of tools designed for that vs. hand editing just temperatures. EnergyPlus doesn't cycle HVAC equipment like they are actually controlled but you can gather fraction fo runtime from your measured data can compare that to fraction runtime for simulated equipment.

For comparing simulated output like fraction runtime or zone temperatures the time series objective function measure is a good option. This compares a CSV file of measured data against SQL output variable from EnergyPlus. We used this measure to calibrate our ORNL FRP model to measured building data. For part of this we had one week of with HVAC systems off and known plug load and occupancy fed in as schedules, with zone temperatures as the objective function. That is described in this paper. It may be good idea to run the simulation for a week or more before you need it bout just request SQL data or evaluate it against your test week. This is important if it has a lot of mass or has been un-conditioned prior to the test period. Hope this helps. that measure is not documented outside of the paper. We do plan to add more documentation and tests for it.