Scalable Bayesian optimization for model calibration: Case study on coupled building and HVAC dynamics
Model calibration for building systems is a key step to achieving accurate and reliable predictions that reflect the dynamics of real systems under study. Calibration becomes particularly challenging when integrating building and HVAC dynamics, due to large-scale, nonlinear, and stiff underlying differential algebraic equations. In this paper, we describe a framework for calibrating multiple parameters of cou-pled building/HVAC models using scalable Bayesian optimization (BO), whose advantages include global optimization without requiring gradient information, and data-efficiency. The proposed methodology is improved online via two additional steps: domain tightening and domain slicing, both of which leverage the learned calibration cost function to reduce the search space volume and dimension, respectively. We demonstrate effectiveness of the proposed algorithm by simultaneously calibrating 17 parameters (in-cluding emissivities, heat transfer coefficients, and thickness of walls/floors) of a Modelica model of joint building and HVAC dynamics, with 2 weeks worth of building data. This high-dimensional calibration task is solved via our proposed scalable BO calibration method, and yields parameters that are > 90% accurate with < 1000 model simulations; additionally, the outputs of the final calibrated model on unseen testing data complies with standard ASHRAE calibration guidelines. (c) 2021 Elsevier B.V. All rights reserved.
WOS:000706262800006
2021-12-15
253
111460
REVIEWED