Forest fire risk prediction is essential for building a forest fire defense system. Ensemble learning methods can avoid the problem of difficult model selection for disaster susceptibility prediction and can significantly improve modeling accuracy. This study introduces a stacking ensemble learning model for predicting forest fire risks in Yunnan Province by integrating various data types, such as meteorological, topographic, vegetation, and human activity factors. A total of 70,274 fire points and an equal number of randomly selected nonfire points were used to develop the model, with 70% of the data allocated for training and the remaining 30% for testing. The stacking model combined four diverse machine learning methods: random forest (RF), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and multilayer perceptron (MLP). We evaluated the model’s predictive performance using metrics like accuracy, area under the characteristic curve (AUC), and fire density (FD). The results demonstrated that the stacking fusion model exhibited remarkable accuracy with an AUC of 0.970 on the test set, significantly surpassing the performance of individual machine learning models, which had AUC values ranging from 0.935 to 0.953. Furthermore, the stacking fusion model effectively captured the maximum fire density in extremely high susceptibility areas, demonstrating enhanced generalization capabilities.

With the development of society and the advancement of technology, the application of electricity in modern life has become increasingly widespread. However, the risk of electrical fires has also significantly increased. This paper thoroughly investigates the causes, classifications, and challenges of electrical fires in special environments, and summarizes advanced detection and extinguishing technologies. The study reveals that the causes of electrical fires are complex and diverse, including equipment aging, improper installation, short circuits, and overloading. In special environments such as submarines, surface vessels, and aircraft, the risk of electrical fires is higher due to limited space, dense equipment, and difficult rescue operations. This paper also provides a detailed analysis of various types of electrical fires, including cable fires, electrical cabinet fires, transformer fires, battery fires, data center fires, and residential fires, and discusses their characteristics and prevention and control technologies. In terms of detection technology, this paper summarizes the progress of technologies such as arc detection, video detection, and infrared thermography, and emphasizes the importance of selecting appropriate technologies based on specific environments. Regarding extinguishing technologies, this paper discusses various means of extinguishing, such as foam extinguishing agents, dry powder extinguishing agents, and fine water mist technology, and highlights their advantages, disadvantages, and applicable scenarios. Finally, this paper identifies the limitations in the current field of electrical fire prevention and control, emphasizes the importance of interdisciplinary research and the development of advanced risk assessment models, and outlines future research directions.

As society rapidly evolves and urbanization accelerates, fire safety has become an increasingly pressing concern

This work discusses the effect of the placement of the thermal release mechanism with temperature sensing element on the actuation time of the fire rolling shutter. The fire rolling shutter is one of the most important devices that can prevent the spread of the fire and smoke by automatically falling when there is a fire in the building. The reliability of the fire rolling shutters can directly affect the fire safety level of the buildings. The thermal release mechanism with temperature sensing element is independent of the linkage control system and is an effective measure to ensure the landing of the fire rolling shutters when other control measures fail. Different fire models are established to simulate the temperature distribution by considering three location factors, including the distance between the release mechanism and the fire, the ceiling and the side wall. Suggestions are proposed based on the numerical simulation results. The thermal release mechanism with temperature sensing element should be installed in the locations that near the fire load and the ceiling, but be far away from the corner and the side wall. The proper placement can improve the efficiency and reliability of the release mechanisms.

The disequilibrium in different fire causes was analyzed by considering province, month and hour. The spatial and temporal distribution characteristic between fire causes and influence factors were investigated based on Gini coefficient and Lorenz curve, which provided a novel method and perspective to measure the disequilibrium of fire occurrence. Seven causes of fires were chosen and there existed an obvious disequilibrium between different provinces. It illustrated that playing with fire tend to keep an extreme disequilibrium both in month and hour. The high-risk factors were explored and these factors should be paid more attention. The results enhance understanding of the spatial and temporal pattern of fires and contribute to the improvement of fire prevention and management strategies.

SiO₂ aerogel cement composites (SACCs) are promising for building insulation, but how their residual thermal performance evolves after high-temperature exposure remains unclear, limiting fire protection assessment. In this study, SACC specimen with aerogel contents of 0%, 5%, 7%, and 10% were heat-treated at 400, 600, 700, 800, and 1000 °C. After cooling, their post-exposure thermal performance and microstructure were characterized via mass loss, density, thermal conductivity, MIP, and SEM. Results obtained at room temperature showed that with increasing treatment temperature, thermal conductivity first decreases and then increases, reaching a minimum after 700 °C treatment for the A7 specimens (from 0.092 to 0.063 W/(m·K)). Microstructural analysis of cooled specimens revealed that this non-monotonic behavior arises from three heat-induced changes: the cement matrix, aerogel aggregates, and the interfacial gap between them. After treatment at 700 °C, the gap corresponds to a Knudsen number of 0.01–0.02, entering the slip-flow regime. Combined with the low thermal conductivity of the cement matrix, this yields the best insulation. After treatment at 800 °C and above, the gap exceeded 60 μm, shifting heat transfer to the continuum regime and reducing insulation capacity. A thermal conductivity prediction model based on these post-exposure mechanisms agreed well with the experimental results.