While cold regions in China experience harsh winters, their summers also present significant overheating challenges in atrium-style buildings due to excessive solar gain. This study investigates the thermal environment of a non-ventilated atrium educational building located in a cold region of China, with tests conducted throughout the four seasons. The findings indicate that the atrium temperature is 1.7 °C, 1.1 °C, and 1.7 °C lower than that of the inner corridor during spring, autumn, and winter, respectively, but 0.6 °C higher in summer. From 7:00 a.m. to 10:00 p.m. on summer days, north-facing rooms with horizontal shading are 0.5 °C warmer than those facing south. A retrofit strategy that combines ventilated atrium shading with north-facing vertical shading is proposed, leading to a 0.7 °C reduction in atrium temperature and a 9.4% decrease in summer air conditioning energy consumption. Additionally, this study develops a retrofit framework for existing buildings, encompassing scope definition, diagnostics, strategy formulation, and evaluation to support high-quality renovations.

Against the backdrop of global energy conservation and emission reduction strategies, university teaching buildings are key targets for carbon emission management due to their multifunctional characteristics and distinct usage patterns. To address the limitations of conventional prediction models in capturing dynamic characteristics of hourly operational carbon emission, a teaching building cluster at a university in Guangzhou was selected as the case study. Hourly occupancy, meteorological data, and electricity consumption data were integrated. A baseline long short-term memory (LSTM) model and seven hybrid deep learning models were developed. The predictive performance of the models was systematically compared using the one-year and two-year training datasets. The results show that the data scale significantly affects the model’s applicability. Under the one-year dataset, the Weekly Seasonal Naive Baseline Model achieves the best overall reference performance. In the deep learning models, the CNN-ATT-LSTM model performs best in trend fitting and absolute error control, while the CNN-LSTM model shows more balanced performance in relative error control and repeated training stability. At the same time, the CNN-ATT-LSTM model achieves a better balance between prediction performance and computational efficiency. Under the two-year dataset, the CNN-LSTM model exhibits optimal overall performance across all prediction accuracy indicators and achieves the most favorable balance between prediction accuracy and computational cost. Further analysis indicates that model performance does not monotonically increase with increasing model complexity. Therefore, the model structure should be matched with the available data basis and specific application requirements. The research results can serve as a basis for model selection for short-term carbon emission prediction in university teaching buildings and provide technical support for improving carbon emission management in these buildings.