Search Results (161)

Internal combustion engine vehicles damage the environment and public health by emitting toxic fumes, such as CO₂ or CO and other trace compounds. The use of electric cars helps to reduce the emission of pollutants into the environment due to the use of batteries with no direct and local emissions. However, accidents of battery electric vehicles pose new challenges, such as thermal runaway. Such accidents can be serious and, in some cases, may result in uncontrolled overheating that causes the battery pack to spontaneously ignite. In particular, the most dangerous vehicles are heavy goods vehicles (HGVs), as they release a large amount of energy that generate high temperatures, poor visibility, and respiratory damage. This study aims to determine the potential consequences of large BEV fires in road tunnels using computational fluid dynamics (CFD). Furthermore, a comparison between a BEV and an ICEV fire shows the differences related to the thermal and the toxic impact. Furthermore, the adoption of a longitudinal ventilation system in the tunnel helped to mitigate the BEV fire risk, keeping a safer environment for tunnel users and rescue services through adequate smoke control. Full article

Low-altitude drones, which are unimpeded by traffic congestion or urban terrain, have become a critical asset in emergency rescue missions. To address the current lack of emergency rescue data, UAV aerial videos were collected to create an experimental dataset for action classification and localization annotation. A total of 5082 keyframes were labeled with 1–5 targets each, and 14,412 instances of data were prepared (including flight altitude and camera angles) for action classification and position annotation. To mitigate the challenges posed by high-resolution drone footage with excessive redundant information, we propose the SlowFast-Traffic (SF-T) framework, a spatio-temporal sequence-based algorithm for recognizing traffic accident rescue actions. For more efficient extraction of target–background correlation features, we introduce the Actor-Centric Relation Network (ACRN) module, which employs temporal max pooling to enhance the time-dimensional features of static backgrounds, significantly reducing redundancy-induced interference. Additionally, smaller ROI feature map outputs are adopted to boost computational speed. To tackle class imbalance in incident samples, we integrate a Class-Balanced Focal Loss (CB-Focal Loss) function, effectively resolving rare-action recognition in specific rescue scenarios. We replace the original Faster R-CNN with YOLOX-s to improve the target detection rate. On our proposed dataset, the SF-T model achieves a mean average precision (mAP) of 83.9%, which is 8.5% higher than that of the standard SlowFast architecture while maintaining a processing speed of 34.9 tasks/s. Both accuracy-related metrics and computational efficiency are substantially improved. The proposed method demonstrates strong robustness and real-time analysis capabilities for modern traffic rescue action recognition. Full article

Lightning strikes pose a significant risk during outdoor activities. The connection between conventionally used rescue blankets in alpine emergencies and the risk of lightning injury is unclear. This experimental study investigated whether rescue blankets made of aluminum-coated polyethylene terephthalate increase the likelihood of lightning injuries. High-voltage experiments of up to 2.5 MV were conducted in a controlled laboratory setting, exposing manikins to realistic lightning discharges. In a balanced test environment, two conventionally used brands were investigated. Upward leaders frequently formed on the edges along the fold lines of the foils and were significantly longer in crumpled rescue blankets (p = 0.004). When a lightning strike occurred, the thin metallic layer evaporated at the contact point without igniting the blanket or damaging the underlying plastic film. The blankets diverted surface currents and prevented current flow to the manikins, indicating potentially protective effects. The findings of this experimental study suggest that upward leaders rise from the edge areas of rescue blankets, although there is no increased risk for a direct strike. Rescue blankets may even provide partial protection against exposure to electrical charges. Full article

With the global increase in maritime activities, the frequency of maritime accidents has risen, underscoring the urgent need for faster and more efficient search and rescue (SAR) solutions. This study presents an intelligent unmanned aerial vehicle (UAV)-based maritime rescue system that combines GPS-driven dynamic path planning with vision-based dual-target detection and tracking. Developed within the Gazebo simulation environment and based on modular ROS architecture, the system supports stable takeoff and smooth transitions between multi-rotor and fixed-wing flight modes. An external command module enables real-time waypoint updates. This study proposes three path-planning schemes based on the characteristics of drones. Comparative experiments have demonstrated that the triangular path is the optimal route. Compared with the other schemes, this path reduces the flight distance by 30–40%. Robust target recognition is achieved using a darknet-ROS implementation of the YOLOv4 model, enhanced with data augmentation to improve performance in complex maritime conditions. A monocular vision-based ranging algorithm ensures accurate distance estimation and continuous tracking of rescue vessels. Furthermore, a dual-target-tracking algorithm—integrating motion prediction with color-based landing zone recognition—achieves a 96% success rate in precision landings under dynamic conditions. Experimental results show a 4% increase in the overall mission success rate compared to traditional SAR methods, along with significant gains in responsiveness and reliability. This research delivers a technically innovative and cost-effective UAV solution, offering strong potential for real-world maritime emergency response applications. Full article

The growth of maritime transportation has been accompanied by a gradual increase in accident rates, drawing greater attention to the critical issue of man-overboard incidents and drowning. Traditional maritime search-and-rescue (SAR) methods are often constrained by limited efficiency and high operational costs. Over the past few years, drones have demonstrated significant promise in improving the effectiveness of search-and-rescue operations. This is largely due to their exceptional ability to move freely and their capacity for wide-area monitoring. This study proposes an enhanced SR-DETR algorithm aimed at improving the detection of individuals who have fallen overboard. Specifically, the conventional multi-head self-attention (MHSA) mechanism is replaced with Efficient Additive Attention (EAA), which facilitates more efficient feature interaction while substantially reducing computational complexity. Moreover, we introduce a new feature aggregation module called the Cross-Stage Partial Parallel Atrous Feature Pyramid Network (CPAFPN). By refining spatial attention mechanisms, the module significantly boosts cross-scale target recognition capabilities in the model, especially offering advantages for detecting smaller objects. To improve localization precision, we develop a novel loss function for bounding box regression, named Focaler-GIoU, which performs particularly well when handling densely packed and small-scale objects. The proposed approach is validated through experiments and achieves an mAP of 86.5%, which surpasses the baseline RT-DETR model’s performance of 83.2%. These outcomes highlight the practicality and reliability of our method in detecting individuals overboard, contributing to more precise and resource-efficient solutions for real-time maritime rescue efforts. Full article

With the increasing acceptance of green, low-carbon, and sustainable development principles, the rising number of new energy vehicles (NEVs) has raised public concern over traffic safety risks associated with these vehicles. To assist traffic management authorities in efficiently allocating rescue resources, this paper proposes a severity prediction method for the new energy vehicle accidents based on Cost-sensitive Fuzzy XGBoost (CFXGBoost). First, chi-square filtering and wrapper methods are used to extract 20 key features strongly cor-related with accident severity. Then, A fuzzy neural network is employed to combine fuzzy inference results with original features, forming an extended feature set. Moreover, These features are used as inputs to the XGBoost model for severity prediction of the new energy vehicle traffic accidents. Finally, the proposed approach is validated using traffic accident datasets from multiple provinces and cities. Results show that the FXGBoost model achieves a prediction accuracy of 0.92 and outperforms other models in terms of precision, recall, and F1 score, demonstrating its effectiveness in accurately predicting the severity of NEV-related traffic accidents. Full article

In this study, a Hybrid Maritime Risk Assessment Model (HMRA) integrating automated machine learning (AML) and deep learning (DL) with hydrodynamic and Monte Carlo simulations (MCS) was developed to assess maritime accident probabilities and risks. The machine learning models of Light Gradient Boosting (LightGBM), XGBoost, Random Forest, and Multilayer Perceptron (MLP) were employed. Cross-validation of model architectures, calibrated baseline configurations, and hyperparameter optimization enabled predictive precision, producing generalizability. This hybrid model establishes a robust maritime accident probability prediction framework through a multi-stage methodology that ensembles learning architecture. The model was applied to İzmit Bay (in Türkiye), a highly jammed maritime area with dense traffic patterns, providing a complete methodology to evaluate and rank risk factors. This research improves maritime safety studies by developing an integrated, simulation-based decision-making model that supports risk assessment actions for policymakers and stakeholders in marine spatial planning (MSP). The potential spill of 20 barrels (bbl) from an accident between two tankers was simulated using the developed model, which interconnects HYDROTAM-3D and the MCS. The average accident probability in İzmit Bay was estimated to be 5.5 × 10⁻⁴ in the AML based MCS, with a probability range between 2.15 × 10⁻⁴ and 7.93 × 10⁻⁴. The order of the predictions’ magnitude was consistent with the Undersecretariat of the Maritime Affairs Search and Rescue Department accident data for İzmit Bay. The spill reaches the narrow strait of the inner basin in the first six hours. This study determines areas within the bay at high risk of accidents and advocates for establishing emergency response centers in these critical areas. Full article

We analyse data from a rescue database collected at the Innsbruck Alpenzoo (Tyrol, Austria). The sample covers 33 years (1988–2020), and more than 5250 wild birds from 145 species originating from Innsbruck and the surrounding Inn Valley, one of the most densely populated areas in Europe. Both, the total number of birds as well as the number of bird species yearly admitted have increased since 1988. Orphaned nestlings and victims of glass collisions were the most common reasons for admission and responsible for the increase. Species’ susceptibility to accidental causes increased with regional abundance and degree of urbanisation. More urbanised species are characterised by a high proportion of nestlings and juveniles in the sample. The seasonal patterns of deliveries in these species show a peak in the late breeding season, and young birds are particularly susceptible to glass collisions and cat attacks. The species list also includes regionally rare wetland, upland and forest breeders and foreign migrants. Such species show a high proportion of admissions in autumn and collisions with windows play a greater role for short-distance migrants. Our data also suggest that small birds (<15 g body mass) are more likely to collide with glass panes than larger species. In conclusion, our data suggest that basically all bird groups and species are at least occasionally affected by human structures and activities in urbanised landscapes but support the notion that juveniles and migrants are more prone for accidents due to the lack of experience with anthropogenic structures in new areas. Full article

Fire is one of the most serious threatening conditions that endanger the safety of human life and building property. Religious buildings, where activities such as ritual incense burning and parishioner worship are conducted year-round, suffer from high fire risks and incomplete coverage of fire protection facilities, which have led to the frequent occurrence of fire accidents in ancient religious buildings around the globe. This study focuses on fire reconstruction and flame-retardant research for double-roofed ancient Buddhist buildings, addressing a gap in fire protection research for ancient religious buildings, particularly those with unique double-roofed structures. A systematic fire simulation method integrating building information modeling (BIM) and computational fluid dynamics (CFD) is proposed. This approach not only accurately models the complex structures of ancient buildings but also simulates fire and smoke spread paths, providing a scientific basis for fire warnings and firefighting strategies. Firstly, the double-roofed ancient Buddhist building is modeled according to its size through building information modeling (BIM). Secondly, the building modeling is revised, and the fire hazard is modeled based on computational fluid dynamics (CFD). Thirdly, the smoke and temperature sensors for fire warning and sprinkler systems for flame retardant are set. Finally, the fire and smoke spread paths are simulated for determining the location for installing the warning sensor and providing valuable fire rescues strategy. Based on simulations, a fire warning system using smoke and temperature sensors, along with a sprinkler-based flame retardant system, is designed. This integrated design significantly enhances the fire prevention and control capabilities of ancient buildings, reducing the occurrence of fire accidents. By simulating fire and smoke spread paths, the optimal locations for sensor installation are determined, and valuable fire rescue strategies are provided. This simulation-based analytical method greatly improves the precision and effectiveness of fire prevention and control. Experiments validate the flame-retardant and fire warning capabilities of the proposed method, demonstrating its practical application value in protecting ancient buildings from fire. The method offers new insights and technical support for fire protection in religious ancient buildings. Full article

As the final crucial line of defense against well blowout accidents, the implementation of the relief well scheme facilitates the swiftest possible rescue of the target well, thereby minimizing economic losses and enhancing production efficiency. Among them, the relative distance and orientation determination between the relief well and the accident well becomes the key to the rapid intersection and connection of the two wells. However, the inaccessibility of the accident well due to safety factors makes it challenging to accurately determine the relative position of the two wells. The current injection method has now been proved to be able to achieve relief well detection, and some signal processing methods are necessary for optimizing the detection performance. In the process of signal characterization, analytical methods show many advantages due to their direct relevance to the principles of electromagnetic theory. To provide an effective theoretical basis for the optimization of relief well detection methods, this paper proposes an analytical solution to the corresponding electromagnetic field distribution for current injection detection scheme. Based on the principle of the current injection detection, the signal model is constructed, and the electromagnetic distribution expression is deduced. Then, the adaptability of the analytical solution obtained under different physical parameters is investigated. Moreover, the reliability and validity of the proposed analytical solution are validated by comparing the results with those obtained from finite element numerical simulations, thus providing a theoretical basis for subsequent signal processing and method optimization. Full article

Every time the proper functioning of the vehicles must be guaranteed, as well as safety and efficiency. To achieve this, some expensive solutions are used, with few connectivity options and that fail to meet consumer demand. This paper presents a low-cost hardware system for the design of a real-time communication protocol between the electronic control unit (ECU) of a vehicle and a remote server based in a embedded system. A dual tone multi-frequency (DTMF) approach is implemented, so error codes (DTCs) are always available on a unit equipped with this system. The vehicle-to-infrastructure (V2I) communication protocol through voice channels is provided by cellular technology infrastructure, in which primary information is shared to monitor vehicles. With real-time data transmission, communication is established through a voice phone call between the vehicle’s ECU and the destination server, communicating the DTC codes. The system shows that the communication protocol has an effectiveness of 78.23%, which means that with the use of 2G technology, which is active and operating in many regions, it allows the information with the data to be received by the receiving user. Through this implemented system, it is ensured that if a vehicle suffers an accident or stops due to a mechanical failure in a region where there is no cellular technology coverage, information or a message can be sent so that through communication the rescue can be carried out using an cellular technology coverage. Full article

It is particularly important to improve the emergency rescue response ability of heavy-haul railways to ensure the safety of personnel and the efficiency of material transportation. The current research has achieved some results for multi-objective material dispatching, but it does not consider the impact of accident response level and material type on material dispatching scheme. In this study, a heavy-haul railway in China was selected as the research object. By designing a dual-objective material scheduling model, an optimal material scheduling scheme was obtained, and the optimal solution was solved by a non-dominated sorting genetic algorithm (NSGA-II). Under the condition of keeping the station unchanged and ensuring that the total amount of materials remained unchanged, an optimization scheme of relief material reserves that match the risk characteristics of the line is proposed. The results show that, based on the utility theory, the minimum distance of the improved dual-objective material dispatching is reduced by 34.8% (single accident point) and 62.99% (multiple accident points), and the total distance of material dispatching is reduced by 37.92% and 70.57%, respectively, indicating that the optimized reserve scheme can effectively shorten the response time and improve the rescue efficiency. The material reserve optimization scheme for emergency rescue stations proposed in this study has important reference value for improving the emergency rescue efficiency of heavy-haul railways. Full article

Unmanned Aerial Vehicles (UAVs) play an extremely important role in real-time object detection for maritime emergency rescue missions. However, marine accidents often occur in low-visibility weather conditions, resulting in poor image quality and a lack of object detection samples, which significantly reduces detection accuracy. To tackle these issues, we propose GD-Det, a low-data object detection model with high accuracy, specifically designed to handle limited sample sizes and low-quality images. The model is primarily composed of three components: (i) A lightweight re-parameterization feature extraction module which integrates RepVGG blocks into multi-concat blocks to enhance the model’s spatial perception and feature diversity during training. Meanwhile, it reduces computational cost in the inference phase through the re-parameterization mechanism. (ii) A cross-scale gather-and-distribute pyramid module, which helps to augment the relationship representation of four-scale features via flexible skip fusion and distribution strategies. (iii) A decoupled prediction module with three branches is to implement classification and regression, enhancing detection accuracy by combining the prediction values from tri-level features. (iv) We also use a domain-adaptive training strategy with knowledge transfer to handle low-data issues. We conducted low-data training and comparison experiments using our constructed dataset AFO-fog. Our model achieved an overall detection accuracy of 84.8%, which is superior to other models. Full article

Towing operations are widely applied in various fields such as maritime accident rescue, assisting large vessels entering and exiting ports, and transporting large ocean platforms. Tugboats and the towed objects form a complex multi-body system connected by flexible cables, and during operations, they are subjected to the effects of complex marine environmental loads. Current research focuses on using numerical simulations and model tests in water tanks to study the motion response of towed objects and cables under the action of environmental loads. There is a lack of research that combines the mechanical response and structural strength with the load conditions of towing operations. Taking cables as an example, most studies focus on the mechanical properties of cables without considering the impact of towing conditions. After reviewing the literature, this paper summarizes the shortcomings of the existing research and points out several potential research directions in the field of towing: the mechanical response of cables during the initial stage of towing, experiments on towing by multiple tugboats, research on composite fiber cables using experimental and finite element simulation methods, and structural optimization of components related to towing operations. Full article

In recent years, firefighter accidents and people injured by the collapse of steel structures during a fire have occurred frequently, which has attracted the attention of the National Emergency Management Department and the Fire and Rescue Bureau. It is urgent to carry out research on early warning systems for building collapse during a fire. Existing early warning methods mainly use characteristic parameters such as temperature, vibration, and structural deformation. Due to the complexity of an actual fire, it is difficult to accurately predict the critical temperature of fire−induced instability in columns and the failure mode after the instability, and there are deviations in the collapse warnings. In this study, changes in ultrasonic transverse and longitudinal wave velocities at high temperatures are used to monitor the stiffness degradation of columns in fire in real time and improve the accuracy of early warning systems. In this study, four common collapse modes of portal frames are obtained by using the results of parametric numerical analysis. According to key displacements and the displacement rates of simple key measuring points, the elastic modulus threshold of a three−level early warning for portal frame collapse with different collapse modes is obtained. Combined with an ultrasonic experiment, the theoretical relationships between the transverse and longitudinal wave velocities and the elastic modulus of steel at high temperatures are verified, and the relationship between the transverse and longitudinal wave velocities and the overall damage of the portal frame is further constructed; then, a new early warning method for portal frame stability during a fire is proposed. Based on the change in wave velocity, a three-level early warning index for predicting portal frame stability during a fire is determined. When the collapse mode of a portal frame is an overall inward collapse, transverse and longitudinal wave velocities are reduced to 2635 m/s and 5308 m/s, respectively. At a second-level warning, they are reduced to 2035 m/s and 4176 m/s, respectively. At 1504 m/s and 3030 m/s, respectively, third-level warnings are issued. This research shows that the real−time monitoring of wave velocities provides an effective way for early warning systems to identify structural collapse. The proposed early warning method can be used as a quick and efficient early warning system for the collapse of portal frames during a fire, and its accuracy and applicability are verified by experiments. Full article