Carbon Performance Evaluation of Urban Buildings Using Machine Learning-Based Energy Models
- 235 Downloads
Because of the large number of buildings and the complexity of influencing factors in urban building energy analysis, the computational cost of large-scale energy models for urban buildings is high. Hence, this paper implements a machine learning-based approach to assess carbon emissions of urban buildings. A campus in China is used as a case study to demonstrate this approach, including four steps: automation of construction of engineering-based energy models in a geographic information system (GIS) environment, construction of a matrix of input and output data, creating machine learning models, and sensitivity and carbon emission analysis of urban buildings. The results indicate that this method can effectively assess carbon emissions for urban buildings with different energy-saving measures based on machine learning energy models. The accuracy of machine learning models should be carefully evaluated before implementing them in carbon emission analysis for urban buildings.
KeywordsMachine learning Building energy model Urban buildings Predictive performance Carbon performance
This research was supported by the National Natural Science Foundation of China (No. 51778416) and the Key Projects of Philosophy and Social Sciences Research, Ministry of Education (China) (contract No. 16JZDH014).
- 3.Tian, W., Rysanek, A., Choudhary, R., Heo, Y.: High resolution energy simulations at city scale. In: IBPSA Building Simulation Conference (BS2015), Hyderabad, India, 7–9 Dec 2015Google Scholar
- 8.MOC, GB50189-2005: Energy Conservation Design Regulation for Public Buildings. Ministry of Construction (MOC) of P.R.China, China Planning Press (2005) (in Chinese)Google Scholar
- 9.Liu, Y.: Energy Saving of Urban Buildings Based on 3D Geographic Information System. Master Degree Master Thesis, Tianjin University of Science and Technology, Tianjin, China (2018)Google Scholar
- 10.DOE: EnergyPlus V9.0.1, October 2018, Department of Energy, USA (2018)Google Scholar
- 11.R Core Team: R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria (2017). URL https://www.R-project.org/. Accessed 10 Dec 2018
- 13.TURCTC, DB29-153-2010: Tianjin design standard for energy efficiency of public buidings. Tianjin Urban-Rural Construction and Transportation Commission (TURCTC). (in Chinese) (2010)Google Scholar