Model Creation and Simulation
Building the model
Start with geometry: spaces versus thermal zones. A thermal zone is the control volume served by one thermostat, so zones should follow HVAC zoning and solar/orientation exposure rather than architectural room names. Envelope inputs - U-factors including thermal bridging, SHGC and visible transmittance, infiltration - must match the design documents. Internal loads (people, lighting, plug loads, process loads) and their schedules drive results and must be traceable to a source.
HVAC and controls
Models are assembled from air loops, zone equipment, plant loops, and controls. Understand auto-sizing versus hard-sizing: autosize is appropriate for design development, but a final model should be reviewed so that oversized or undersized equipment does not distort part-load performance. Setpoints, economizers, deadbands, and supply air temperature reset all change results materially.
Baseline and proposed
For 90.1 Appendix G work, the baseline model keeps prescribed envelope, lighting, and schedules while the proposed model reflects the actual design; HVAC types are assigned by the system mapping table. A frequent error is altering baseline inputs that must remain fixed.
Traps
Watch for convergence warnings and unmet load hours being ignored rather than resolved, weather files mismatched to the project location, missing fans and pumps, and schedules applied inconsistently between baseline and proposed. Do not treat a completed run as a validated model.
Rule of thumb
Run a quick sanity check before detailed review: compare simulated EUI against a published benchmark for the same building type and climate. A 2–3× discrepancy usually indicates a units, schedule, or HVAC configuration error, not a modeling subtlety.