Sensitivity analysis methods for building energy models: Comparing computational costs and extractable information

Menberg, Kathrin; Heo, Yeonsook; Choudhary, Ruchi

doi:10.1016/j.enbuild.2016.10.005

Cited by 149 publications

(79 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In accordance with the equations of the KOH framework, we refer to the quantity of interest as model output (y), to contour state variables as inputs (x), and to all types of unknown system characteristics as parameters, of which those chosen for calibration are denoted as θ (Figure 2). We perform a sensitivity analysis on the heat pump model applying Morris method to identify the most influential unknown model parameters using the load side outlet temperature of the heat pump as quantity of interest (Morris 1991;Menberg, Heo, and Choudhary 2016). Of the 12 uncertain parameters in the heat pump model, we focus on the four most influential parameters (θ): load side fluid specific heat [kJ kg −1 K −1 ] (θ 1 ), rated cooling capacity of the heat pump [kJ h −1 or W] (θ 2 ), source side flow rate in the heat pump [kg h −1 ] (θ 3 ) and load side flow rate in the heat pump [kg h −1 ] (θ 4 ) ( Figure 2), which together explain most of the variance in the model output caused by the uncertain parameters.…”

Section: Calibration Parameters Of the Energy Supply Modelmentioning

confidence: 99%

Influence of error terms in Bayesian calibration of energy system models

Menberg

Heo

Choudhary

2018

Journal of Building Performance Simulation

Self Cite

View full text Add to dashboard Cite

Calibration represents a crucial step in the modelling process to obtain accurate simulation results and quantify uncertainties. We scrutinize the statistical Kennedy & O'Hagan framework, which quantifies different sources of uncertainty in the calibration process, including both model inputs and errors in the model. In specific, we evaluate the influence of error terms on the posterior predictions of calibrated model inputs. We do so by using a simulation model of a heat pump in cooling mode. While posterior values of many parameters concur with the expectations, some parameters appear not to be inferable. This is particularly true for parameters associated with model discrepancy, for which prior knowledge is typically scarce. We reveal the importance of assessing the identifiability of parameters by exploring the dependency of posteriors on the assigned prior knowledge. Analyses with random datasets show that results are overall consistent, which confirms the applicability and reliability of the framework.

show abstract

Section: Calibration Parameters Of the Energy Supply Modelmentioning

confidence: 99%

Influence of error terms in Bayesian calibration of energy system models

Menberg

Heo

Choudhary

2018

Journal of Building Performance Simulation

Self Cite

View full text Add to dashboard Cite

show abstract

“…The method of Morris belongs to the class of One-factor-At-a-Time (OAT) design (only one parameter changes values between consecutive simulations) and is suitable when the number of input factors are so large that other variance-based approaches are computationally prohibitive (Saltelli et al, 2008). Therefore, it is a common technique for carrying out sensitivity analysis in building energy models (Heo et al, 2012;De Wit and Augenbroe, 2002;Tian, 2013;Menberg et al, 2016;Kristensen and Petersen, 2016). The main advantage of the Morris method is its relatively lower computation cost as compared to other global sensitivity analysis methods, making it particularly well-suited for use with building energy models where the number of uncertain parameters is high.…”

Section: Current Bayesian Calibration Methods 221 Sensitivity Analysmentioning

confidence: 99%

Bayesian calibration of building energy models with large datasets

Chong

Lam

Pozzi

et al. 2017

Energy and Buildings

101

View full text Add to dashboard Cite

“…Moreover, Menberg et al . adopted analytical approaches and complicated mathematical methods to study the impacts of parameters on building energy model . However, these methods may require additional IT software due to computational difficulties.…”

Section: Introductionmentioning

confidence: 99%

“…To study the relation between input and outputs of a building energy model, first and second order sensitivity analyses were performed [18]. Moreover, Menberg et al adopted analytical approaches and complicated mathematical methods to study the impacts of parameters on building energy model [19]. However, these methods may require additional IT software due to computational difficulties.…”

Section: Introductionmentioning

confidence: 99%

Impact of building design parameters precision on heating and cooling load calculations

Abasnezhad

Soltani

Markarian

et al. 2018

Env Prog and Sustain Energy

View full text Add to dashboard Cite

Energy efficiency of buildings is of paramount importance due to limited fossil resources, environmental issues, and increasing energy demand. Thus, in recent years building energy analysis using various programs has experienced a rapid growth. However, these programs are expensive and time consuming. In this study, mechanical engineering building of Tabriz University located in northwest of Iran is analyzed to attain a simplified building model that can reduce the computational time and effort. Therefore, 10 models are considered with distinctive design accuracies. EnergyPlus software has been employed to perform energy analysis of the building. The findings show that building materials, internal walls and zoning of the building are determining factors in thermal loads of the building. However, shadings, number of occupants, and electrical appliances have negligible impact. Thus, it can be concluded that the model in which a detailed envelope design is considered and interior walls, shades, electrical appliances, and duration of occupancy are omitted, has the optimum performance in terms of time saving and accuracy of load calculation. Moreover, to gain more reliable results, it is recommended to consider zones in model No.5. The difference between natural gas and electricity consumption of the first and last models is 20,302 m3 (2200$) and 1157 kWh (90$) that indicates the significance of proper HAVC sizing. It is worth mentioning that all monetary units are presented in 2018 U.S. dollar. © 2018 American Institute of Chemical Engineers Environ Prog, 38: 741–749, 2019

show abstract

Sensitivity analysis methods for building energy models: Comparing computational costs and extractable information

Cited by 149 publications

References 32 publications

Influence of error terms in Bayesian calibration of energy system models

Influence of error terms in Bayesian calibration of energy system models

Bayesian calibration of building energy models with large datasets

Impact of building design parameters precision on heating and cooling load calculations

Contact Info

Product

Resources

About