Greetings!
A colleague and I have been working through some energy modeling in BEopt for a multi-family housing project in Bozeman, MT and have been running into issues with unmet loads. Initially we were thinking the issues were rooted in the mini-split we had selected. To test this, we ran some simulations with different mini-splits and other more conventional mechanical systems but the issues persisted, see below -
Proposed design w/ mini-split and electric baseboard - 1374 hours of unmet loads for the whole building year round (126 heating, 1248 cooling). Of that, 1261 hours of the unmet cooling loads were from unit 4. There were 14 hours of unmet heating loads for unit 4.
Autosizing mini-split for cooling loads - Increased unmet cooling loads by 1 hour
Air-source heat pump - (#4) No change in cooling loads, 153 hour increase in unmet heating loads (Also tested #8 and 9 and they did even worse)
Conventional mech. systems, Central AC (#4) + Electric furnace (#2) w/ no baseboard - Increased unmet cooling loads by 107 hours, 671 hour increase in unmet heating loads
Conventional mech. systems, Central AC (#4) + Electric boiler (#2) w/ no baseboard - Increased unmet cooling loads by 107 hours, 156 hour increase in unmet heating loads
Alternate mini-split (#16) - Increased unmet cooling loads by 22 hours, 5 hour increase in unmet heating loads
I also ran a B-10 benchmark for reference - 478 hours of unmet loads for the whole building year round (236 heating, 242 cooling). Of that, 185 hours of the unmet cooling loads were from unit 4. There are no unmet heating loads for unit 4.
Through the troubleshooting process we discovered that the vast majority of unmet cooling loads were concentrated to one of four units in the building (#4). Unit 4 is located on the second floor and has exterior walls on 2 sides. There is a unit with the same footprint directly below it on the ground floor (#2). It also shares a roof with the other two multi-floor units (#1 and #3) located on both sides of units 2 and 4. As a general observation it also seems that the more insulation we add, the greater the unmet cooling loads become. Conceptually this makes sense but we get confused when the mechanical systems have been set to autosize based on cooling demand and they are not actually doing so. We are also modeling in a heating dominated climate and some of the numbers for unmet cooling loads would be almost the entirety of the cooling season, leading us to believe the cooling system is not working correctly.
Has anyone else seen these issues? If so, what solutions and/or workarounds can you offer? Is is possible that there is an issue with the autosizing algorithm in BEopt?
Any insight would be greatly appreciated!
