Argonne National Laboratory

Upcoming Events

A Model Calibration and Validation Framework for Building Performance Simulation Applications under Uncertainty

Series 
CEEESA/Grid Seminar
Presenter 
Qi Li, Georgia Institute of Technology
October 13, 2017 11:30AM to 12:30PM
Location 
Building 362, Room F108
Type 
Seminar

Abstract: The calibration and validation of building performance simulation (BPS) models remain an important subject of study in fault detection, operations management, and retrofit analysis. This study develops a new model calibration and validation framework that builds upon uncertainty propagation using detailed measurements and inverse modeling using Bayesian inference. This framework also introduces probabilistic accuracy metrics to assess model prediction accuracy.

Two case studies are provided to demonstrate the framework's effectiveness. The first is a joint work with Argonne on uncertainty quantification of empirical validation experiments. A new validation methodology is proposed to validate a simulation model under uncertainty, in which the validation criteria build upon the introduced probabilistic accuracy metrics. Uncertainty propagation based on synthetic measurements is applied, which effectively improves prediction agreement and reduces the risk of accepting invalid simulation outcomes.

The second case is to determine the appropriate model form and metering data for a hypothetical intervention analysis of an existing building with hydronic heating on the Cambridge, UK, campus. Different sets of metering data are used to calibrate a physical building simulation model, and the result indicates the superiority of Bayesian inference in exploiting the value of data, the necessity of electricity monitoring under uncontrolled conditions, and the potential for daily metering data for calibration in real building performance management practice.

Bio: Qi Li is a graduate research assistant in the High-Performance Buildings Lab at Georgia Institute of Technology. His research focuses on engineering-based data-driven building energy modeling; urban-scale building energy management; and statistical analysis, optimization, and data visualization in sustainable building design, operation, and retrofit practice. He completed his Ph.D. at the School of Architecture at Georgia Institute of Technology, M.S. in statistics at Georgia Institute of Technology, and B.Eng. in building technology at Tsinghua University in China.