Paper Conference

Proceedings of ASHRAE/IBPSA-USA Building Simulation Conference


Impact of Modeler Decisions on Simulation Results

Pamela Berkeley, Philip Haves, Erik Kolderup
University of California Berkeley, Berkeley, CA
Lawrence Berkeley National Labs, Berkeley, CA
Kolderup Consulting, San Francisco, CA

Abstract: The user generating the model for Building Energy Simulation programs is responsible for a variety of decisions that affect simulation results. A study was conducted to determine the magnitude and nature of the modelers role in simulation results. 12 professional energy modelers were given identical plans for a commercial building and instructed to complete a model of the building in eQUEST. Results for total yearly electrical consumption varied by −11% to +104% and results for gas consumption varied by −61% to +1535%. A one at a time factor analysis was performed on the participant modeling decisions with a baseline model created from modeling best practices. The results of this analysis indicate that decisions related to equipment power levels caused the most variation in electricity predictions, while HVAC decisions caused the greatest deviation from the baseline model predictions for both gas and electricity consumption. Future work should investigate a broad range of buildings and modelers and additionally explore the possibility of reducing modeler variation by modifying simulation program user interfaces.
Pages: 203 - 210