A Bayesian Network model for predicting cooling load of commercial buildings
- 235 Downloads
Cooling load prediction is indispensable to many building energy saving strategies. In this paper, we proposed a new method for predicting the cooling load of commercial buildings. The proposed approach employs a Bayesian Network model to relate the cooling load to outdoor weather conditions and internal building activities. The proposed method is computationally efficient and implementable for use in real buildings, as it does not involve sophisticated mathematical theories. In this paper, we described the proposed method and demonstrated its use via a case study. In this case study, we considered three candidate models for cooling load prediction and they are the proposed Bayesian Network model, a Support Vector Machine model, and an Artificial Neural Network model. We trained the three models with fourteen different training data datasets, each of which had varying amounts and quality of data that were sampled on-site. The prediction results for a testing week shows that the Bayesian Network model achieves similar accuracy as the Support Vector Machine model but better accuracy than the Artificial Neural Network model. Notable in this comparison is that the training process of the Bayesian Network model is fifty-eight times faster than that of the Artificial Neural Network model. The results also suggest that all three models will have much larger prediction deviations if the testing data points are not covered by the training dataset for the studied case (The maximum absolute deviation of the predictions that are not covered by the training dataset can be up to seven times larger than that of the predictions covered by the training dataset). In addition, we also found the uncertainties in the weather forecast significantly affected the accuracy of the cooling load prediction for the studied case and the Support Vector Machine model was more sensitive to those uncertainties than the other two models.
KeywordsBayesian Network model cooling load prediction training dataset uncertainties
Unable to display preview. Download preview PDF.
This research was supported by the U.S. National Science Foundation under award number IIS-1633338. This research was also supported by the U.S. Department of Defense under the ESTCP program. The authors thank, Ana Carolina Laurini Malara, Marco Bonvini, Michael Wetter, Mary Ann Piette, Jessica Granderson, Oren Schetrit, Rong Lily Hu and Guanjing Lin for the support provided through the research.
- ASHRAE (2012). ASHRAE Handbook—HVAC Systems and Equipment. Atlanta, GA, USA: American Society of Heating, Refrigerating, and Air-Conditioning Engineers.Google Scholar
- Birdsall B, Buhl WF, Ellington KL, Erdem AE, Winkelmann FC (1990). Overview of the DOE-2 Building Energy Analysis Program, Version 2. 1D. Technical Report LBL-19735-Rev.1.Google Scholar
- Chen Y, Xu P, Chu Y, Li W, Wu Y, Ni L, Bao Y, Wang K (2017). Short-term electrical load forecasting using the Support Vector Regression (SVR) model to calculate the demand response baseline for office buildings. Applied Energy, 196: 659–670.Google Scholar
- EIA (2016). The State Energy Data System. The U.S. Energy Information Administration. Availabe at http://www.eia.gov/ state/seds. Accessed 9 Mar 2016.Google Scholar
- Huang S, Zuo W (2014). Optimization of the Water-cooled Chiller Plant System Operation. Paper presented at the ASHRAE/IBPSAUSA Building Simulation Conference, Atlanta, GA, USA.Google Scholar
- Huang S, Zuo W, Sohn MD (2016b). A Bayesian network model for predicting the cooling load of educational facilities. Paper presented at the ASHRAE/IBPSA-USA Building Simulation Conference, Salt Lake City, UT, USA.Google Scholar
- Huang S, Malara ACL, Zuo W, Sohn MD (2016c). A Bayesian network model for the optimization of a chiller plant’s condenser water set point. Journal of Building Performance Simulation, doi: 10.1080/19401493.2016.1269133.Google Scholar
- Hughes JT, Domínguez-García AD, Poolla K (2015). Virtual battery models for load flexibility from commercial buildings. Paper presented at the 48th Hawaii International Conference on on System Sciences, Hawaii, USA.Google Scholar
- Kashiwagi N, Tobi T (1993). Heating and cooling load prediction using a neural network system. Paper presented at the International Joint Conference on Neural Networks, Nagoya, Japan.Google Scholar
- Kim SH (2011). Development of robust building energy demand-side control strategy under uncertainty. PhD Thesis, Georgia Institute of Technology, USA.Google Scholar
- Klein SA, Duffie JA, Beckman WA (1976). TRNSYS—A transient simulation program. ASHRAE Transactions, 82: 623–633.Google Scholar
- Krati M (2016). Energy Audit of Building Systems: An Engineering Approach, 2nd edn. New York: CRC Press.Google Scholar
- Li X, Ding L, Li Y, Xu G, Li J (2010a). Hybrid genetic algorithm and support vector regression in cooling load prediction. Paper presented at the 3rd International Conference on Knowledge Discovery and Data Mining, Phuket, Thailand.Google Scholar
- Li X, Ding L, Lv J, Xu G, Li J (2010b). A novel hybrid approach of KPCA and SVM for building cooling load prediction. Paper presented at the 3rd International Conference on Knowledge Discovery and Data Mining, Phuket, Thailand.Google Scholar
- O’Neill Z (2014). Development of a probabilistic graphical energy performance model for an office building. Paper presented at the 2014 ASHRAE Annual Meeting, Seattle, WA, USA.Google Scholar
- Sakawa M, Ushiro S, Kato K, Ohtsuka K (1999). Cooling load prediction in a district heating and cooling system through simplified robust filter and multi-layered neural network. Paper presented at the IEEE International Conference on Systems, Man, and Cybernetics, Tokyo, Japan.Google Scholar
- Schaul T, Bayer J, Sun Y, Felder M, Sehnke F, Rückstieß T, Schmidhuber J (2010). PyBrain. Journal of Machine Learning Research, 11: 743–746.Google Scholar
- U.S. DOE (2014). Buildings Energy Data Book. U.S. Department of Energy. Available at https://catalog.data.gov/dataset/buildingsenergy-data-book. Accessed 15 May 2016.Google Scholar