Fire Technology最新文献

This paper presents a new evacuation model for fast and affordable simulations of evacuation based on Togawa’s theory for multi-compartment configurations. The aim is to track the evacuee’s path and to estimate the congestion (or the queues) behind each doorway at each time step to model the evacuation process. In this approach, only two parameters drive the formation of congestion, namely the maximum out-coming people flux and the width of the doorway. For a real application, such as evacuation in a building or a boat, a geometrical configuration is considered by a “tree structure” where each doorway is connected to the others up to the main exit. The originality of this paper is in proposing a theoretical expression for the people flux feeding the congestion for people which are located just behind a given doorway. Moreover, this contribution proposes various new experimental tests to qualify and to validate the proposed model. All experimental data (146 evacuation exercises) are available in an open access database for further uses. In this communication, a sensitivity analysis is proposed on a single deck evacuation of the RMS Titanic (the best documented ship for its geometry) with 1126 people. This analysis demonstrates that, between the free walk speed and the maximum out-coming people flux per length of doorway, this latter variable is the most influential parameter of the present model, accounting for 22% of variations in evacuation time. The model has been applied to estimate evacuation times for generic Ro–Ro ships, to test some existing alternatives to abandon a ship and to propose some new perspectives to optimize the evacuation.

To address the challenges of early forest fire smoke image recognition, including false alarms and missed reports caused by interference in complex environments, an enhanced model, named MB-YOLO, is proposed based on the YOLOv8 Nano (YOLOv8n) architecture for efficiently recognizing forest fire smoke. Firstly, to overcome detection failures of low-concentration smoke in complex backgrounds, the original Path Aggregation Network (PAN) is replaced with a bi-directional feature pyramid network (BiFPN). This substitution not only enhances multi-scale feature extraction but also simplifies the network structure, reducing the number of parameters. Secondly, to address false detections caused by cloud and mist interference, the C2f_MLCA module is developed. This module integrates a lightweight Mixed Local Attention mechanism (MLCA) into the bottleneck of the gradient flow module C2f, thereby enhancing smoke feature extraction. Lastly, to reduce sensitivity to positional offsets of small smoke targets, the Complete Intersection over Union (CIoU) loss is replaced with Inner-DIoU loss. This new loss function computes loss with auxiliary bounding boxes, accelerating convergence speed and enhancing accuracy for small smoke targets. The effectiveness of the algorithm is validated with a curated dataset containing small smoke targets, unclear backlighting, and cloud and mist interference. Experimental results demonstrate that our model achieves a mean Average Precision (mAP) of 80.1%, a frame rate of 60.6 Frames Per Second (FPS), with a total of 1.09 million parameters and 7.1 billion floating-point operations per second (FLOPs). This model offers high detection accuracy, fewer parameters, and lower GFLOPs, facilitating accurate real-time monitoring of forest fires in complex environments and all-weather conditions.

The propagation of smoke and hot gases in mechanically ventilated nuclear compartments has been highlighted as one of the main issues of significance. It may lead to the failure of several systems such as clogging of filters located in the ventilation network or electrical devices. To address this issue, the continuous improvement of the predictive capability of existing models with regards to liquid pool fires is of high importance. Computational fluid dynamics (CFD) is widely used for fire simulations. It is worth noting that most pool fire simulations in open atmosphere, under-ventilated and mechanically ventilated compartments have relied on pre-defined/prescribed fuel mass loss rate (MLR) or heat release rates (HRR) from correlations or experimental data when available. Therefore, the prediction of fuel MLR and HRR based on the specific actual fire conditions rather than prescribed data, remains a key development area for the fire community. The present work aims to provide some contribution and advances on this issue. Building on existing liquid evaporation models, the study develops an approach which in then implemented in an in-house version of the CFD code FireFOAM in which a mechanical ventilation model has been embedded, to predict the fuel MLR in both open atmosphere and mechanically ventilated compartments. Validations of the implemented model includes comparison with experimental fuel MLR and previous studies that made use of correlations and experimental data. The results show acceptable fuel MLR predictions with reasonable accuracy and provide further insights into fire behaviour in mechanically ventilated compartments.

Wildfires pose an escalating threat to critical infrastructure, particularly transmission lines, leading to severe power outages and significant economic impacts. While existing studies have primarily focused on static risk assessment methods, this research introduces a novel dynamic risk assessment framework that addresses the rapidly evolving nature of wildfire dynamics through advanced data assimilation techniques, utilizing a real-world wildfire case study. Unlike previous approaches that rely on single-parameter updates or static fire line predictions, our framework integrates observational data into the wildfire simulation tool FARSITE using an ensemble transform Kalman filter, enabling multi-parameter updates that significantly enhance the predictive accuracy of fire line positions and their associated uncertainties. Furthermore, a Monte Carlo simulation-based approach is developed to dynamically calculate wildfire arrival probabilities, combined with a robust quantitative framework for assessing transmission line failure likelihood under fire scenarios. The fire line intensity, determined under the worst-case scenario principle, serves as the input for the quantitative assessment framework. By integrating wildfire arrival probabilities and transmission line failure risks, this study provides a comprehensive and dynamic risk assessment tool, offering a transformative perspective on managing the interface between wildfires and critical infrastructure.

The high thermal conductivity of steel, combined with its rapid degradation in mechanical properties with increasing temperature, makes it vulnerable to fire. Fire protection materials are effectively designed to control the temperature rise within steel members. This paper is a companion to a previous numerical analysis study on protected square hollow section (SHS) steel columns using thermally enhanced gypsum-based mortars. It offers a more detailed numerical investigation into the thermal performance of different gypsum-based mortar compositions used as a passive fire protection material for different types of steel columns. Firstly, finite element models for SHS steel columns were developed and verified against data from previous fire resistance tests. Then, a parametric study was conducted to explore how factors like fire protection thickness and composition, cross-section (square, rectangular, and H-shaped sections), steel tube thickness, column slenderness, and applied load level (serviceability load states) affect their fire performance under the ISO-834 standard fire curve. Comparisons were made between numerical results and current design methods from Eurocodes. It was observed that existing design methods excessively underestimate the actual fire resistance of protected columns, particularly for class-4 cross-sections especially when mortars with highest thermal insulation capacity are used. Moreover, the thermal properties of fire protection mortars should be considered in the structural steel temperature prediction as a function of temperature during fire conditions. Based on the study’s findings, modifications to current design methods for predicting the temperature evolution of columns as a function of the cross-sections and fire protection compositions, were presented with enhanced accuracy. These proposed modifications can potentially contribute to future development in Eurocode and improved fire resistance predictions.

Modern heating, ventilation, and air conditioning (HVAC) systems are complex, interconnected systems optimised to be energy efficient. Damper-optimised demand-controlled ventilation systems (DCV) minimise energy consumption by using a dedicated control unit that calculates the optimal fan speed based on room sensors and the feedback from all DCV dampers, which each measures the airflow rate and adjusts its damper angle accordingly. In buildings that do not use a compartmentation strategy in the event of a fire, it is crucial that the ventilation system is pressurised and provides balanced ventilation in order to prevent smoke from spreading via the ventilation system and to avoid creating pressure imbalances, which may impair evacuation. In the present study, two full-scale fire tests from a series of 14 tests in a mock-up building equipped with a damper-optimised DCV system are presented, and the ventilation system’s performance during the fire is assessed. The tests revealed various failure mechanisms caused by heat exposure, leading to individual damper uncontrolled opening or closing or the building management system losing contact with all dampers. Furthermore, it was shown that the failure of individual dampers and the gradual clogging of the extraction filter can affect the pressure balance in other parts of the building outside the fire room and increase the risk of smoke spreading through the ventilation ducts.

The growing incidence of wildfires, intensified by changing climate patterns, poses risks to human lives and the environment, leading to catastrophic impacts on agricultural and forest ecosystems. Consequently, timely wildfire detection becomes imperative to implement effective mitigation strategies. This article presents a new forest fire image detection technique to address a class imbalance problem using ResNet-18 multi-resolution features and kernel extreme learning machine (KELM). Shallow and deep layer ResNet-18 features are extracted and fused to create a comprehensive feature set that represents local and global characterization of the forest fire image data. The multi-resolution feature fusion effectively captures lower-level visual patterns and complex and abstract representations of the input image. The fused feature set is subsequently input into a kernel extreme learning machine, which effectively handles nonlinear data patterns for binary classification tasks like fire detection. However, the performance of the KELM heavily relies on its hyperparameters, which are optimized using the Newton–Raphson-Based Optimizer (NRBO) algorithm. The hyperparameters fine-tuning process ensures that the KELM operates with optimal settings, ultimately enhancing the accuracy and reliability of the fire detection process. The proposed algorithm is evaluated using two publicly available databases, Forest Fire and Flame, with a detection accuracy of 97.88% and 99.88%, respectively. Moreover, the contribution of each feature to the model’s predictions to interpret decisions is elaborated using SHAP (SHapley Additive exPlanations).

Fire accidents are a critical design condition for load-bearing elements in general. Among others, ordinary glass and composite glass materials are even more susceptible to fire and require the use or definition of specific test protocols, simulation strategies, performance indicators and validation methods. In this paper, the structural performance of a full-scale composite timber-glass composite wall (consisting of a perimetral timber frame and a double thin insulating glass unit (IGU)) under the effects of sustained mechanical loads (25 kN/m, as in a typical two-story building) and fire exposure is investigated based on a standard test furnace. The mechanical concept uses a laminated system that can cover an area of up to 3.2 × 2.7 square meters, with a relatively low thickness (63.52 mm for the double insulating glass unit (IGU), including cavity). A great advantage to evaluate the potential and critical points of the composite timber-glass composite system comes from experimental and finite element (FE) thermomechanical investigations. A pilot test is being conducted on a prototype prefabricated timber-glass module, which is expected to function as an efficient load-bearing system in buildings, withstanding the typical mechanical loads from normal or extreme actions, but also providing adequate resistance to fire accidents. The laboratory investigation was carried out on the basis of conventional recommendations for the experimental assessment of building components in the event of fire, with the main focus on estimating fire resistance. It has been shown that the overall load-bearing capacity and the corresponding fire resistance are mainly determined by the intrinsic properties of the glass components, which may need to be protected or optimized to ensure adequate residual capacity.

Fluorinated substances are widely used in the different components of the lithium-ion battery cell, such as electrode binders, electrolyte, additives and separator materials. To date, most studies regarding the fluorinated contaminations from lithium-ion battery fires are focused on the gases formed, whereas the solids produced are not as well characterized. Here, we present an experimental study investigating the occurrence of per- and polyfluoroalkyl substances, in soot and particulates formed after thermal runaway in lithium-ion battery cells. Per- and polyfluoroalkyl substances were detected in every battery cell test performed in this study. The concentration of per- and polyfluoroalkyl substances ranged between 20 to 130 ng/g_soot. Extrapolation of data gives an estimated release of 10 to 60 µg of per- and polyfluoroalkyl substances per kg battery cells. Among the 22 per- and polyfluoroalkyl substances analyzed, perfluorobutanesulfonic acid and perfluorobutanoic acid were found in the highest concentrations for all samples. Interestingly, perfluorooctanoic acid was detected in all tests, in concentrations ranging between 0.05 to 0.62 ng/g_soot. These findings are of importance not only for the purpose of decontamination after thermal runaway events, but also when it comes to the lithium-ion battery recycling processes.