As distributed energy systems become increasingly prevalent, residential energy systems (RES) equipped with photovoltaics (PV) face significant challenges in maintaining supply-demand balance due to power output fluctuations. This necessitates short-term PV power prediction methods that effectively balance accuracy and deployment cost. To address this issue, this paper proposes a novel short-term PV power prediction approach based on low-cost ground-based sky image sequences: the 3DCNN-DLinear model. The method leverages fisheye camera-captured sky images to extract spatiotemporal features via a three-dimensional convolutional neural network (3DCNN), and integrates a lightweight time-series model, DLinear, to enable efficient prediction. The proposed model was evaluated using real-world data collected in Changping District, Beijing, China. A comparative analysis involving six mainstream time-series models confirmed that DLinear achieved the lowest overall prediction error. Further experiments demonstrated that the 3DCNN-DLinear model reduced RMSE by 49.28%, 9.56%, and 8.82% for 30-, 60-, and 90-minute prediction tasks, respectively, compared to the baseline 3DCNN-LSTM model. Additionally, the study examined the contribution of sky image data to prediction accuracy, revealing significant improvements under varying conditions. Notably, RMSE was reduced by 40.4% and 30.5% under sunny and cloudy conditions, respectively, for the 60-minute task. Overall, the proposed method offers an effective and economically viable solution to improve the predictive performance and intelligent scheduling of RES.

Indoor airflow distribution significantly influences temperature regulation, humidity control, and pollutant dispersion, directly impacting thermal comfort and occupant health. Accurate and efficient prediction of airflow fields is essential for optimizing ventilation systems and enabling real-time control. However, existing computational approaches for dynamic ventilation are computationally intensive and have limited generalization capabilities. This study leverages the Fourier neural operator (FNO), a method rooted in operator learning and Fourier transform principles, to develop a three-dimensional (3D) airflow simulation model capable of predicting velocity and its components. The model was trained using 200 s of sinusoidal ventilation data (amplitude: 0.4) and evaluated under diverse air supply patterns, including sinusoidal (amplitude: 0.8), intermittent, and stepwise periodic ventilation. Additionally, the model’s performance was assessed with low-resolution training data and further tested for recursive prediction accuracy. Results reveal that the FNO method achieves high accuracy, with a mean square error of 9.906 × 10^–5 for sinusoidal amplitude 0.8 and 4.004 × 10^–5 over 400 time steps for sinusoidal, intermittent, and stepwise conditions. Further evaluations, including tests on low-resolution training data and recursive prediction, were conducted to examine the model’s resolution invariance and assess its performance in iterative forecasting. These findings demonstrate the FNO method’s potential for robust, efficient prediction of 3D unsteady airflow fields, providing a pathway for real-time ventilation system optimization.

Winter poses a high risk for spreading infectious respiratory diseases, particularly in classrooms, which are known hotspots for cross-infection. The health hazards in classrooms lacking mechanical ventilation systems often go unnoticed. To address this issue, we studied the risk of respiratory infection transmission in such environments during winter by assessing the spread of contaminants using computational fluid dynamics (CFD). We evaluated four common airflow setups (split-type air conditioner, open door, open window, and both door and window open) in classrooms without mechanical ventilation. Our findings indicate that while split air conditioners provide optimal thermal comfort, they significantly increase exposure risk. Conversely, simply opening a door can effectively balance thermal comfort with a reduced exposure risk—particularly when the infected individual is near the door, leading to a minimal 0.01% average intake fraction. Furthermore, under varying ventilation scenarios, the sensitivity of exposure risk to location changes of infected individuals differs significantly. Specifically, when a split-type air conditioner is used, occupant exposure is largely unaffected by changes in the location of the infected individual. However, the exposure risk becomes highly sensitive to location changes of infected individuals when a door or window is used for ventilation. Strategic positioning of the infected individual can decrease indoor exposure risk by up to 94% with the door open and 67% with the window open. Additionally, when the door and window are both open, the dependency of occupant exposure on the location of the infected individual decreases. In this case, the exposure risk for indoor occupants is low, regardless of the position of the infected individual.

As infectious respiratory diseases are highly transmissible through the air, researchers have improved traditional total volume air distribution systems to reduce infection risk. Multi-vent module-based adaptive ventilation (MAV) is a novel ventilation type that facilitates the switching of inlets and outlets to suit different indoor scenarios without changing ductwork layout. However, little research has evaluated MAV module sizing and air velocity selection, both related to MAV system efficiency in removing contaminants and the corresponding level of protection for occupants in the ventilated room. Therefore, the module-source offset ratio (MSOR) is proposed, based on the MAV module size and its distance from an infected occupant, to inform selection of optimal MAV module parameters. Computational fluid dynamics simulations illustrated contaminant distribution in a two-person MAV equipped office. Discrete phase particles modelled respiratory contaminants from the infected occupant, and contaminant concentration distributions were compared under four MAV air distribution layouts, three air velocities, and three module sizes considered using the MSOR. Results indicate that lower air velocities favour rising contaminant levels, provided the ventilation rate is met. Optimal contaminant discharge can be achieved when the line of outlets is located directly above the infected occupant. Using this parameter to guide MAV system design, 85.7% of contaminants may be rendered harmless to the human body within 120 s using the default air vent layout. A more appropriate supply air velocity and air vent layout increases this value to 91.4%. These results are expected to inform the deployment of MAV systems to reduce airborne infection risk.

Indoor thermal comfort is essential as it improves living standards. Activity scenarios of personnel are in the process of a dynamic change. In most interior spaces with fixed working stations, people directly blown by cold air have a poor thermal experience. Therefore, to meet the differentiated environmental demands, one challenge is to explore novel ventilation strategies to satisfy the changing environmental needs. An adaptive strategy, multi-vent module-based adaptive ventilation (MAV), was designed to increase the adjustability of air distribution and better adapt to variable demands. A classroom was chosen as a representative model with multiple scenarios during its use. Simulations were conducted to verify the three-level control effect of MAV on improving the thermal environment. The results revealed that different vent solutions create different airflow patterns and thermal environments, which can be matched to the scenarios. The scale for ventilation efficiency No. 4, which measured the influence scope of supply air, was used to evaluate the zoning division control in MAV. The space under the charge of a concerned MAV module showed a higher SVE4 than that at other zones. This implied that the zoning division can be effectively implemented. Thermal comfort measured using the air diffusion performance index, predicted mean vote, and draught rate showed that the application of MAV is better than that of the fixed MV mode, and the discomfort experienced when exposed to cold air can be avoided. It is believed that these results will help extend the research of adaptive ventilation strategies.