Prediction and Comparative Study of Energy Consumption in Typical Primary School Classroom Units under Different Cooling Control Strategies

This study investigates the impact of different cooling control strategies on energy consumption in a typical primary school classroom unit, based on a fixed occupancy schedule. Two strategies were evaluated: one relying solely on air conditioning and the other prioritizing natural ventilation. To account for climatic variation, three representative cities were selected according to China's climate zoning: Beijing (cold region), Shanghai (hot summer and cold winter), and Guangzhou (hot summer and warm winter). By comparing the energy use intensity (EUI) across these strategies and climate zones, the study explores how design variables influence simulated energy consumption under different environmental conditions. A machine learning model was subsequently developed to predict cooling EUI, enabling rapid estimations across various scenarios. The results reveal significant variations in EUI among the different climate zones. Furthermore, discrepancies in simulation accuracy between the two control strategies were observed across climatic conditions. Feature importance analysis highlights that the influence of design variables on EUI varies depending on both the climate zone and the control strategy. This study provides valuable insights for evaluating and optimizing energy performance in primary school buildings. It also establishes a practical “simulation-prediction-response” closed-loop framework to support demand-side energy planning and decision-making.