For example, if I run the model for the coolingDesignDay to get the breakdown of heat gain components, it seems these results could be proportionally representative of what is contributing most greatly (by %) to cooling energy in general? Similarly, I would then run another simulation on the heatingDesignDay to get a representation of the heat loss breakdown.
The problem is that the design day simulations aren’t proportionally representative of annual energy. It helps to think about this statistically. Design day periods represent the ~0.4%-ile or ~99.6-ile of your annual dry bulb temperatures for cooling and heating respectively[1]. By definition that means it’s a rare event. For Gaussian distributions, anything that is approximately three standard deviations from the mean (0.03%-ile/99.7%-ile) can safely be considered outliers. Another common statistical technique, (for non-Gaussian distributions) is to consider anything outside the 25th and 75th percentiles (the interquartile range) as outliers. Regardless of which technique we use, by definition the design days are not representative of the conditions informing your whole building energy.
My suggestion is to work with the actual constituent data to work out your breakdown, rather then try and map different analysis periods to it. The latter method can lead you astray. When I suggested splitting the analysis into two, I meant something like taking the monthly heat balance, and averaging all the months with mechanical cooling into one breakdown, and averaging all the months with mechanical heating into another. I’ve done that in the past to try and distill the monthly data into a representative heating and cooling season, although this may not work if you have shoulder seasons/moderate climates with simultaneous heating and cooling, so there needs to be way to account for that as well (or just omit it).
A related question on ReadEP Results terms: MechVentilation appears when I am running an Ideal Air Loads model but disappears when I add an HVAC system (VAV w/Reheat). How can I account for OA ventilation gains/losses when I have a specific system added?
The component loads aren’t represented when you model specific systems because the addition of HVAC system efficiencies result in mechanical cooling/heating quantities that are less then the sum of the component loads. So it’s more difficult to conceptually represent your breakdown this way. However you can find and account for different component loads (even if they don’t add up) by looking up different EnergyPlus outputs in the RDD file, and adding it your simulation run as a requested output.
[1] As a general question to the LB community, I’ve always been confused about how ASHRAE defines their percentiles. If you bin your dry bulb temperatures from coolest to warmest temperatures, then by convention the 0.4%-ile should correspond to the colder temperatures, aka your heating design day (reading from left to right). But ASHRAE defines this as the hotter temperatures, which means they’re reading their probability distribution from right to left. What am I missing?