The limitation of that Hydra example is that doesn’t model the coupling between a building foundation and the ground, and thus it’s simulated temperatures will be to extreme for actual ground temperatures around and underneath your greenhouse.
To approximate the impact of ground coupling, you can’t just model the ground as a thermal zone, adjacent to building zones due to the following two complications:
- The thermal zone surface conduction calculations model heat transfer in 1-dimension, which thus assumes the entire surface plane is isothermal (temperature is constant). But ground heat transfer needs to be modeled in 2 or even 3 dimensions (since it’s a large, wet, thermal mass that dynamically changes along the footprint of the entire building at different depths).
- Swings in ground temperatures operate on the timescale of months, and even years, whereas EnergyPlus thermal zones (which represents the ground in that example) operates on an hourly timescale.
The most accurate way to do this in EnergyPlus is with Kiva[1], which models dynamic, multidimensional ground/foundation heat flow, and moisture transportation at different timescales.
Unfortunately I don’t think Honeybee has integrated Kiva yet. I’m not exactly sure what Honeybee uses currently to model building foundation heat transfer, although I do know it’s accounting for some of the complexity I mentioned above (i.e computing custom ground temperatures for the deep ground boundary condition). I’ll leave that for someone else to answer.