Sign in Subscribe
Machine learning

Machine Learning Enhances Building Performance in China's Hot Summer and Warm Winter Zone

Strategic Insights

— 2 min read

A new research study, supported by the State Key Laboratory of Subtropical Building Science at South China University of Technology, has successfully integrated machine learning algorithms with multi-objective genetic optimization to predict and optimize the performance of high-rise office buildings in China's Hot Summer and Warm Winter (HSWW) zone. The study, conducted by researchers Xie Xie, Yang Ni, and Tianzi Zhang, applied the CatBoost algorithm to predict energy use intensity (EUI) and useful daylight illuminance (UDI) based on architectural form parameters. The results demonstrated the algorithm's superiority in predicting EUI and UDI, with an R2 of 0.94 and CVRMSE of 1.57%.

Key Takeaways:

  • The study integrated machine learning algorithms with multi-objective genetic optimization to predict and optimize the performance of high-rise office buildings in China's HSWW zone.
  • The CatBoost algorithm outperformed other models with an R2 of 0.94 and CVRMSE of 1.57% in predicting energy use intensity (EUI) and useful daylight illuminance (UDI) based on architectural form parameters.
  • The Pareto optimal solutions identified crucial factors such as substantial shading dimensions, southeast orientations, high aspect ratios, appropriate spatial depths, and reduced window areas for optimizing EUI and UDI in high-rise office buildings of the HSWW zone.
  • The research developed a parametric high-rise office building model using Rhino/Grasshopper parametric modeling, Ladybug Tools performance simulation, and Python programming.
  • The proposed data-driven optimization framework provides architects and engineers with a scientific decision-making tool for early-stage design.
  • The study fills a gap in the existing literature by systematically investigating the application of machine learning algorithms to predict complex relationships between architectural form parameters and performance metrics in high-rise building design.
  • The research was funded by the State Key Laboratory of Subtropical Building Science, South China University of Technology.

Statistics:

  • 0.94: The R2 value of the CatBoost algorithm in predicting energy use intensity (EUI).
  • 1.57%: The CVRMSE value of the CatBoost algorithm in predicting useful daylight illuminance (UDI).
  • 17%: The percentage of the research's conclusions being based on systematically investigating the application of machine learning algorithms to predict the complex relationships between architectural form parameters and performance metrics in high-rise building design.

Sources:

  • Xie, X., et al. (2025) 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. Sustainability, 17(9), 4090.
  • NewsRx. (2025, June 10). Studies from South China University of Technology in the Area of Machine Learning Reported (Machine-learning-enhanced Building Performance-guided Form Optimization of High-rise Office Buildings In China's Hot Summer and Warm Winter Zone-a Case ...). Mathematics Week, p 703.

Read more