| id |
caadria2024_51 |
| authors |
Ye, Jiahong, Shen, Yanting, Huang, Chenyu, Wang, Jinyu, Qu, Rong and Yao, Jiawei |
| year |
2024 |
| title |
Explainable AI and Multi-objective Optimization For Energy Retrofits in Existing Residential Neighborhoods |
| doi |
https://doi.org/10.52842/conf.caadria.2024.1.485
|
| source |
Nicole Gardner, Christiane M. Herr, Likai Wang, Hirano Toshiki, Sumbul Ahmad Khan (eds.), ACCELERATED DESIGN - Proceedings of the 29th CAADRIA Conference, Singapore, 20-26 April 2024, Volume 1, pp. 485–494 |
| summary |
Globally, the carbon emissions of building sector contribute to 40% of all. Therefore, low-carbon retrofits of buildings become pivotal to carbon neutrality. While existing retrofit research focus on individual building without considering whether different neighborhood morphologies would affect the effectiveness of retrofit strategies. Here we propose a framework for optimal retrofitting in diverse urban neighborhoods, balancing energy savings and cost. Based on actual residential neighborhood morphologies, massive parametric sample models have been established and batch simulated before and after the retrofitting for its energy performance. Then, we use XGBoost to establish a predictive model for energy savings, and iGeneS to perform multi-objective optimization. SHapley Additive exPlanations is utilized to reveal the nonlinear contribution of four retrofitting strategies to building retrofit benefits (BRB). Results demonstrate that in different floor area ratio (FAR), the contributions of installing rooftop photovoltaic panels to BRB vary. In neighborhoods with FAR of 1.5-2.8, the contribution of installing photovoltaic panels is not as significant as in other FAR ranges, and its contribution to BRB is not comparable to replacing energy-efficient lights. Moreover, The effectiveness of deep lighting fixture retrofitting may be suboptimal. The proposed framework will offer efficient energy-saving guidance for future residential neighborhood retrofits. |
| keywords |
Residential neighbourhood, Low-carbon retrofits, Urban morphology, Energy consumption, Explainable AI, Multi-objective optimization |
| series |
CAADRIA |
| email |
jiawei.yao@tongji.edu.cn |
| full text |
 file.pdf (4,206,030 bytes) |
| references |
Content-type: text/plain
|
Aruta, G., Ascione, F., Bianco, N., & Mauro, G. M., (2023)
 Sustainability and energy communities: Assessing the potential of building energy retrofit and renewables to lead the local energy transition
, Energy, 282, 128377
|
|
|
|
|
Gustavsson, L., & Piccardo, C., (2022)
Cost optimized building energy retrofit measures and primary energy savings under different retrofitting materials, economic scenarios, and energy supply
, Energies, 15(3), 1009
|
|
|
|
|
Hong, T., Chen, Y., Luo, X., Luo, N., & Lee, S. H., (2020)
Ten questions on urban building energy modeling
, Building and Environment, 168, 106508
|
|
|
|
|
Huang, C., Zhang, G., Yin, M., & Yao, J. (2022)
Energy-driven intelligent generative urban design
, CAADRIA, volume 1, 233-242
|
|
|
|
|
Liu, K., Xu, X., Zhang, R., Kong, L., Wang, W., & Deng, W. (2023)
Impact of urban form on building energy consumption and solar energy potential: A case study of residential blocks in Jianhu, China
, Energy and Buildings, 280, 112727
|
|
|
|
|
Lu, Y., Chen, Q., Yu, M., Wu, Z., Huang, C., Fu, J., ... & Yao, J. (2023)
Exploring spatial and environmental heterogeneity affecting energy consumption in commercial buildings using machine learning
, Sustainable Cities and Society, 95, 104586
|
|
|
|
|
Ma, D., Li, X., Lin, B., Zhu, Y., & Yue, S. (2023)
A dynamic intelligent building retrofit decision-making model in response to climate change
, Energy and Buildings, 284, 112832
|
|
|
|
|
Zhang, Y., Teoh, B. K., Wu, M., Chen, J., & Zhang, L. (2023)
Data-driven estimation of building energy consumption and GHG emissions using explainable artificial intelligence
, Energy, 262, 125468
|
|
|
|
| last changed |
2024/11/17 22:05 |
|
