Machine-Learning-Enhanced Building Performance-Guided Form Optimization of High-Rise Office Buildings in China’s Hot Summer and Warm Winter Zone—A Case Study of Guangzhou DOI Open Access
Xie Xie,

Ni Yang,

Tianzi Zhang

et al.

Sustainability, Journal Year: 2025, Volume and Issue: 17(9), P. 4090 - 4090

Published: May 1, 2025

Given their dominant role in energy expenditure within China’s Hot Summer and Warm Winter (HSWW) zone, high-fidelity performance prediction multi-objective optimization framework during the early design phase are critical for achieving sustainable efficiency. This study presents an innovative approach integrating machine learning (ML) algorithms genetic to predict optimize of high-rise office buildings HSWW zone. By Rhino/Grasshopper parametric modeling, Ladybug Tools simulation, Python programming, this developed a building model validated five advanced mature predicting use intensity (EUI) useful daylight illuminance (UDI) based on architectural form parameters under climatic conditions. The results demonstrate that CatBoost algorithm outperforms other models with R2 0.94 CVRMSE 1.57%. Pareto optimal solutions identify substantial shading dimensions, southeast orientations, high aspect ratios, appropriate spatial depths, reduced window areas as determinants optimizing EUI UDI research fills gap existing literature by systematically investigating application ML complex relationships between metrics design. proposed data-driven provides architects engineers scientific decision-making tool early-stage design, offering methodological guidance similar regions.

Language: Английский

Machine-Learning-Enhanced Building Performance-Guided Form Optimization of High-Rise Office Buildings in China’s Hot Summer and Warm Winter Zone—A Case Study of Guangzhou DOI Open Access
Xie Xie,

Ni Yang,

Tianzi Zhang

et al.

Sustainability, Journal Year: 2025, Volume and Issue: 17(9), P. 4090 - 4090

Published: May 1, 2025

Given their dominant role in energy expenditure within China’s Hot Summer and Warm Winter (HSWW) zone, high-fidelity performance prediction multi-objective optimization framework during the early design phase are critical for achieving sustainable efficiency. This study presents an innovative approach integrating machine learning (ML) algorithms genetic to predict optimize of high-rise office buildings HSWW zone. By Rhino/Grasshopper parametric modeling, Ladybug Tools simulation, Python programming, this developed a building model validated five advanced mature predicting use intensity (EUI) useful daylight illuminance (UDI) based on architectural form parameters under climatic conditions. The results demonstrate that CatBoost algorithm outperforms other models with R2 0.94 CVRMSE 1.57%. Pareto optimal solutions identify substantial shading dimensions, southeast orientations, high aspect ratios, appropriate spatial depths, reduced window areas as determinants optimizing EUI UDI research fills gap existing literature by systematically investigating application ML complex relationships between metrics design. proposed data-driven provides architects engineers scientific decision-making tool early-stage design, offering methodological guidance similar regions.

Language: Английский

Citations

0