TY - GEN
T1 - Efficiency and reliability of bayesian calibration of energy supply system models
AU - Menberg, Kathrin
AU - Heo, Yeonsook
AU - Choudhary, Ruchi
N1 - Funding Information:
This study is supported by EPSRC grant (EP/L024452/1): Bayesian Building Energy Management (B-bem).
Publisher Copyright:
© 2017 Building Simulation Conference Proceedings. All rights reserved.
PY - 2017
Y1 - 2017
N2 - In this study, we examine the efficiency and reliability of a Bayesian calibration setup using temperature point measurements. Hamiltonian Monte Carlo sampling is found to be significantly more efficient with regard to convergence of the posterior distributions, which is assessed using different visual and quantitative measures. The examination of posterior realizations from different data sets and different prior distributions reveals that inference about model parameters is in general quite reliable, while learning about the magnitude of different error terms, such as model discrepancy and random errors, proves to be more difficult. Finally, predictive simulation results based on these inferred posterior distributions are generally in good agreement with measured data.
AB - In this study, we examine the efficiency and reliability of a Bayesian calibration setup using temperature point measurements. Hamiltonian Monte Carlo sampling is found to be significantly more efficient with regard to convergence of the posterior distributions, which is assessed using different visual and quantitative measures. The examination of posterior realizations from different data sets and different prior distributions reveals that inference about model parameters is in general quite reliable, while learning about the magnitude of different error terms, such as model discrepancy and random errors, proves to be more difficult. Finally, predictive simulation results based on these inferred posterior distributions are generally in good agreement with measured data.
UR - http://www.scopus.com/inward/record.url?scp=85095956241&partnerID=8YFLogxK
U2 - 10.26868/25222708.2017.315
DO - 10.26868/25222708.2017.315
M3 - Conference contribution
AN - SCOPUS:85095956241
T3 - Building Simulation Conference Proceedings
SP - 572
EP - 581
BT - 15th International Conference of the International Building Performance Simulation Association, Building Simulation 2017
A2 - Barnaby, Charles S.
A2 - Wetter, Michael
PB - International Building Performance Simulation Association
T2 - 15th International Conference of the International Building Performance Simulation Association, Building Simulation 2017
Y2 - 7 August 2017 through 9 August 2017
ER -