The appropriate design of urban ventilation corridors (VCs) can improve the urban thermal environment, thereby reducing urban energy consumption and promoting sustainable urban development. However, existing research lacks a comprehensive grasp of the characteristics of VCs from multiple dimensions and quantitative analysis of its cooling effect. We propose a novel approach based on data mining to comprehensively consider the morphological and environmental characteristics of VCs and explore the correlation between VC characteristics and the cooling effect. Selecting Nanjing as an example, a comprehensive index system was constructed, the cooling effect of the VC was investigated, and the optimal range of VC index with different underlying surface types was obtained. Results revealed that the cooling effect is closely related to the underlying surface, leading to a temperature difference of up to 5.4 °C. The VC cooling range can vary from 13 to 600 m. The recommended optimal parameter intervals for different VCs were determined. Finally, targeted strategies to alleviate the heat island effect were proposed for different underlying surface types. The study output contributes to the design of VCs, which is of great significance in alleviating the urban heat island effect and promoting sustainable development of cities.

Indoor temperature and relative humidity control in office buildings is crucial, which can affect thermal comfort, work efficiency, and even health of the occupants. In China, fan coil units (FCUs) are widely used as air-conditioning equipment in office buildings. Currently, conventional FCU control methods often ignore the impact of indoor relative humidity on building occupants by focusing only on indoor temperature as a single control object. This study used FCUs with a fresh-air system in an office building in Beijing as the research object and proposed a deep reinforcement learning (RL) control algorithm to adjust the air supply volume for the FCUs. To improve the joint control satisfaction rate of indoor temperature and relative humidity, the proposed RL algorithm adopted the deep Q-network algorithm. To train the RL algorithm, a detailed simulation environment model was established in the Transient System Simulation Tool (TRNSYS), including a building model and FCUs with a fresh-air system model. The simulation environment model can interact with the RL agent in real time through a self-developed TRNSYS–Python co-simulation platform. The RL algorithm was trained, tested, and evaluated based on the simulation environment model. The results indicate that compared with the traditional on/off and rule-based controllers, the RL algorithm proposed in this study can increase the joint control satisfaction rate of indoor temperature and relative humidity by 12.66% and 9.5%, respectively. This study provides preliminary direction for a deep reinforcement learning control strategy for indoor temperature and relative humidity in office building heating, ventilation, and air-conditioning (HVAC) systems.

The development of 5G application technology has led to a rapid expansion in the scale of internet data center rooms and the number of servers. Due to the high heat generation of data center server equipment and the mixing of hot and cold airflows within the rooms, the thermal environment of these rooms fails to meet operational requirements with increasing energy consumption and thermal density. This study utilized the 6SigmaDC software to simulate and analyze the characteristics and existing problems of airflow distribution in a small-scale data center. Based on identified issues with current airflow patterns, two optimization schemes were proposed, analyzing the effects of raised floor height and the closure of aisles on airflow optimization. The return heat index (RHI) was used as an evaluation metric to assess airflow patterns before and after optimization. When the raised floor height was 600 mm, the maximum temperature at the cabinet inlet and outlet were 19.3 °C and 34 °C respectively, which were the lowest, and the RHI value was 0.9622. Compared with unclosed aisles and closed hot aisles, closed cold aisles effectively reduced the cabinet inlet and outlet temperature and increased the RHI. In addition, closed cold aisles increased the air supply temperature from 18 °C to 20 °C, further reducing the energy consumption of the air conditioning system. This study can provide guidance and act as a reference for optimizing airflow design and energy conservation in small data centers.

With increasing global climate change, extreme weather threats to indoor environments are growing. Heatwave events provide essential data for building thermal resilience analysis. However, existing heatwave definition indicators vary widely and lack standardized criteria. To more accurately evaluate indoor overheating risks, this study compared indoor overheating responses under different heatwave definition indicators, considering the temporal disconnect between indoor and outdoor heat conditions. Focusing on Beijing, this study established an indoor–outdoor coupled heatwave evaluation framework using 1951–2021 meteorological data and the heat index as an overheating metric. By analyzing indoor overheating degree and overlap degree to characterize indoor–outdoor correlations, we concluded that different definitions of heatwaves lead to variations in identifications, while multidimensional indicators better capture extreme events. Heatwaves with prolonged duration and high intensity pose greater health risks. Although Beijing’s indoor thermal conditions are generally safe, peak heat indices during summer heatwaves exceed danger thresholds in some buildings, highlighting thermal safety concerns. The metrics for heatwave 6 and heatwave 7 optimally integrate indoor–outdoor characteristics with higher thresholds identifying more extreme events. These findings support the design of building thermal resilience, overheating early warnings, and climate-adaptive electrification strategies.