Search

Although it is widely acknowledged that reinforcement learning (RL) can be beneficial for building control, many RL-based control actions remain unexplainable in the daily practice of facility managers. This paper reports a rule reduction framework using explainable RL to enhance the practicality of the control strategy. First, deep Q-learning was applied to explore the optimal control strategies of a parallel cooling system (ice-based thermal system + geothermal heat pump system) of an existing office building. A set of modularized and interconnected data-driven models was developed using ANNs for pretraining an artificial agent. After exploring the control strategies, the decision-making rules of the agent were reduced using a decision tree. The performance of the reduced-order rule-based control proved comparable to the complex and uninterpretable control strategy of deep Q-learning. The difference in energy savings between the two is marginal at 1.2%. [ABSTRACT FROM AUTHOR]

Data-driven simulation models can be used for optimal control and decision making in building systems only if they are developed accurately and reliably. An unsolved issue pertains to the data collected from BEMS that generally contain errors owing to sensors, malfunctioning systems, and other unknown reasons. These anomalies generally occur in a multidimensional space and cannot be easily detected. Unfortunately, these anomalies can prevent the simulation model from being accurate, reliable, and scalable. This paper presents an automatic anomaly detection method using a support vector data description (SVDD). The data obtained from a compression chiller in a real office building were classified into raw and filtered datasets based on SVDD. Two artificial neural network (ANN) models were developed (ANNraw and ANNSVDD). Based on the comparison between the two ANNs, the ANNSVDD is found to be better than ANNraw in terms of the model reliability and reproducibility. It is also interesting that the model accuracy differences between the two ANN models are marginal. However, the accuracy of the two ANN models can be improved as long as they are tested against filtered data by SVDD. The improvements in accuracy signify the importance of the elimination of these anomalies. [ABSTRACT FROM AUTHOR]

Recently, the issue of parameter identifiability has been highlighted in Bayesian inference of building energy models. Parameter identifiability addresses the correspondence between the observed data and the model parameters, and this analysis verifies whether the model parameters can be estimated using observed data. The authors propose a model selection process for Bayesian inference involving the unidentifiable parameters based on the model evidence. The model evidence is a component of the Bayes theorem, which corresponds to the observation probability (i.e., the goodness of fit) of the observed data for a given model. In this study, nested sampling was used to estimate the model evidence. A case study with the reference office building developed by the US Department of Energy shows the model selection process that uses the model evidence as the evaluation index for the unidentifiable parameter. For comparison with the existing approach of model evaluation, the authors present the results of a comparative analysis between the proposed process and that based on model prediction error (CV(RMSE)). As a result, it was observed that the higher the model evidence in the hypothesis for the unidentifiable parameters, the more the posteriors tend to be similar to the true values. In addition, no significant relationship was observed between the model prediction error and the accuracy of the posterior inference. This indicates that the model evidence can be used as an objective evaluation index for selecting the optimal hypothesis of unidentifiable parameters in the Bayesian inference of building energy models. [ABSTRACT FROM AUTHOR]

Copyright of Building Research & Information is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The purpose of the present study was to investigate the relevance of building thermal performance and characteristics to building energy consumption. This paper reports an energy analysis of 4625 office buildings in Seoul, South Korea, using data from the Korean national building energy database and architectural database. The following four research questions were investigated: (1) Do old buildings consume more energy than new ones? (2) Have strict prescriptive building energy codes contributed to the reduction in energy use intensity (EUI, kWh/m2·year) over the past several decades? (3) What are the characteristics of building energy consumption in terms of season, age, and cooling system (electric chiller vs absorption chiller)? (4) Which factors in the Korean building energy database are relevant to building energy consumption? The analyses revealed that, contrary to common assumptions, new buildings did not always consume less energy than old buildings, and it may be wrong to attribute intensification of prescriptive building energy codes directly to building energy efficiency improvements. In addition, the building characteristics (i.e., district, year built, number of floors, number of elevators, and total floor area) available in the Korean building energy database do not adequately explain building energy consumption, and the existing data collection method needs further improvement. [ABSTRACT FROM AUTHOR]

It is widely acknowledged that Bayesian inference is only beneficial when prior information is properly defined. However, there is no clear rule for prior selection, and it is apparently a matter of subjective selection by the domain expert(s). In other words, because the posterior inference results can vary depending on how the prior is set, a proper definition of the prior is important in terms of objectivity and accuracy for Bayesian inference of building energy models. Hence, the authors suggest a new prior selection method using Dirichlet process Gaussian mixture model (DPGMM) and the likelihood confidence region (hereafter referred to as likelihood CR). The DPGMM is a Bayesian nonparametric clustering technique that optimizes both the cluster shape and the number of clusters. Using the DPGMM, the finite probability distributions that make up the likelihood CR can be estimated, where the distribution with the highest maximum likelihood is applied as the informative prior. In this study, a reference office building of the United States Department of Energy was selected, and a surrogate model was generated using an artificial neural network. Based on a comparison between the authors' suggestion and traditional informative (and/or non-informative) priors by domain experts, the proposed method requires only minimum information about the parameters (min and max) and performs better than the traditional approach [ABSTRACT FROM AUTHOR]