Fire

Forest fires cause extensive environmental damage, making early detection crucial for protecting both nature and communities. Advanced computer vision techniques can be used to detect smoke and fire. However, accurate detection of smoke and fire in forests is challenging due to different factors such as different smoke shapes, changing light, and similarity of smoke with other smoke-like elements such as clouds. This study explores recent YOLO (You Only Look Once) deep-learning object detection models YOLOv9, YOLOv10, and YOLOv11 for detecting smoke and fire in forest environments. The evaluation focuses on key performance metrics, including precision, recall, F1-score, and mean average precision (mAP), and utilizes two benchmark datasets featuring diverse instances of fire and smoke across different environments. The findings highlight the effectiveness of the small version models of YOLO (YOLOv9t, YOLOv10n, and YOLOv11n) in fire and smoke detection tasks. Among these, YOLOv11n demonstrated the highest performance, achieving a precision of 0.845, a recall of 0.801, a mAP@50 of 0.859, and a mAP@50-95 of 0.558. YOLOv11 versions (YOLOv11n and YOLOv11x) were evaluated and compared against several studies that employed the same datasets. The results show that YOLOv11x delivers promising performance compared to other YOLO variants and models. Full article

(1) Background: this study investigates the impact of acid value changes on the thermal degradation and fire risks of palm oil. It emphasizes the need for systematic risk management in food manufacturing and preparation processes to address safety challenges associated with high-temperature operations. (2) Methods: the study employed fire reproduction experiments, fire risk characterization tests, and thermal analyses, including differential scanning calorimetry and thermogravimetric analysis. (3) Result: higher acid values in palm oil significantly reduce smoke points, ignition points, and thermal stability, primarily due to increased free fatty acids and oxidative by-products. These effects are more pronounced in oxidative environments, highlighting the importance of controlling acid value to mitigate fire and thermal risks. (4) Conclusions: this study concludes that increased acid value in palm oil significantly reduces its thermal stability and elevates fire risks due to accelerated oxidation and thermal decomposition. It emphasizes the importance of monitoring acid value and implementing temperature control measures to enhance safety in food manufacturing and cooking processes. Full article

Shifts in global climate patterns can alter animal behavior, including movement and space use. The southwestern United States of America is currently undergoing a period of megadrought, which can have profound consequences on small ectothermic organisms like box turtles. We radiotracked eight adult ornate box turtles (Terrapene ornata) in eastern New Mexico from September 2019 to July 2022, when the environmental conditions transitioned from a dry season with low cumulative precipitation in 2020 to high cumulative precipitation in 2021, followed by a regression to exceptional drought conditions that culminated with a high-intensity wildfire in early 2022. Turtles exhibited greater mean daily movement and were more active in 2021 in comparison to 2020 and 2022. Turtles were least active in 2022, while mean daily movement was comparative to 2020. All turtles in our study exhibited homing behavior after the wildfire, but individual responses varied. While some turtles initially moved out of the burned area and returned within a month, others remained inactive within a small portion of the burned area. The greatest movement was documented in one female turtle following the wildfire, whose home range expanded to seven times the average maximum annual home range size observed among other turtles. Overall, this is the first documentation of T. ornata response to highly altered habitat after high-severity wildfire. Full article

Outdoor fire detection faces significant challenges due to complex and variable environmental conditions. Fiber Optic Distributed Temperature Sensing (FO-DTS), recognized for its high sensitivity and broad monitoring range, provides significant advantages in detecting outdoor fires. However, prediction models trained in laboratory settings often yield false and missed alarms when deployed in complex outdoor settings, due to environmental interferences. To address this issue, this study developed a fixed-power fire source simulation device to establish a reliable small-scale experimental platform incorporating various environmental influences for generating anomalous temperature data. We employed deep learning autoencoders (AEs) to integrate spatiotemporal data, aiming to minimize the impact of outdoor conditions on detection performance. This research focused on analyzing how environmental temperature changes and rapid fluctuations affected detection capabilities, evaluating metrics such as detection accuracy and delay. Results showed that, compared to AE and VAE models handling spatial or temporal data, the CNN-AE demonstrated superior anomaly detection performance and strong robustness when applied to spatiotemporal data. Furthermore, the findings emphasize that environmental factors such as extreme temperatures and rapid temperature fluctuations can affect detection outcomes, increasing the likelihood of false alarms. This research underscores the potential of utilizing FO-DTS spatiotemporal data with CNN-AE for outdoor fire detection in complex scenarios and provides suggestions for mitigating environmental interference in practical applications. Full article

Powder-based fire extinguishing methods are widely used to suppress fires of all kinds efficiently. However, these methods have several drawbacks, the most significant being the large powder residue left behind, which can complicate cleanup and damage sensitive equipment. The present paper investigates reacting flows and develops a methodology for analyzing the interaction of powder particles with fire, addressing both homogeneous and heterogeneous fire inhibition mechanisms. To achieve this, a simplified model was developed using the common principles of the general dynamic equation (GDE) and the population balance equation (PBE) coupled with the reacting flow equations. The model examines the interplay between the initial particles’ diameter and their extinguishing flow rate (concentration), also known as minimal extinguishing concentration (MEC), establishing the relation between the two. Notably, the relation exhibits three different zones, each influenced by different governing mechanisms of combustion inhibition, providing critical insights into optimizing powder-based extinguishing systems. A minimal value of the MEC is found where there is no significant change with the MEC in terms of particle diameter, and the chemical homogeneous mechanism is dominating. The methodology also offers a pathway for finding the maximal extinguishing particle diameter (MED) when the heterogeneous extinguishing mechanism acquires its maximal impact.There is no benefit with a larger particle diameter as it would not practically achieve better extinguishment, but would lead to a potential waste of powder and hence damage equipment. A significant advantage of using extinguishing powders with micro-sized/ultrafine particles is demonstrated where the homogeneous inhibition mechanism becomes predominant. The developed methodology and finding suggest that micro-sized powders are more effective in extinguishing fires, as they offer improved dispersion and reactivity, enhancing the overall efficiency of the fire suppression process. However, considering economic factors such as micron-sized-powder production cost and maintenance may require considering a shift of this set point. Full article

This study investigates the performance of DC corona discharge electrostatic precipitators (ESPs) for NO conversion to increase DeNO_x technologies’ efficiency for small-scale biomass combustion systems. Experiments were conducted using a 5 kW automatic wood pellet domestic heat source with combustion gas treated in a specially designed ESP operated in both positive and negative corona modes, resulting in a reduction in NO concentrations from 130 mg/m³ to 27/29 mg/m³ for positive/negative polarities (at 0 °C and 101.3 kPa). NO conversion efficiency was evaluated across a range of specific input energies (SIEs) from 0 to 50 J/L. The results demonstrate that DC corona ESPs can achieve up to 78% NO reduction, with positive corona demonstrating a greater energy efficiency, requiring a lower SIE (35 J/L) compared to the negative corona mode (48 J/L). A detailed analysis of reaction pathways revealed distinct conversion mechanisms between the two modes. In positive corona, dispersed active species distribution led to more uniform NO conversion, while negative corona exhibited concentrated reaction zones with about 20% higher ozone production. The reactions involving O and OH radicals were more important in positive corona, whereas ozone-mediated oxidation dominated in negative corona. The research results demonstrate that ESP technology with DC corona offers a promising, energy-efficient solution for NO_x control in small-scale combustion systems. Full article

The purpose of this study was to compare heart rate recovery (HRR) after a maximal treadmill (MAX-TM) and three-minute all-out (3MT) test between firefighters (FF) and a control (CON) group. Nine male CON and nine male FF participants completed height (m), weight (kg), body fat percent (BF%), normalized handgrip (GRIP_NORM, kg/kg), and MAX-TM with direct gas analysis to capture aerobic capacity (VO_2PEAK, mL/kg/min). A shuttle-sprint 3MT was used to measure critical velocity (CV, m/s) and D′ (m). Non-linear models determined HR decay (HRR_τ), HR asymptote (HR_∞), and HR amplitude (HR_amp). Two-way GROUP (FF vs. CON) by TEST (MAX-TM vs. 3MT) repeated measures ANOVAs indicated a significant TEST (F = 7.004, p = 0.018) effect on HR_amp. When divided by VO_2PEAK classification (FITNESS), a significant TEST effect was observed (F = 7.661, p = 0.014) on HR_amp. VO_2PEAK was significantly related to CV (r = 0.583, p = 0.011), GRIP_NORM (r = 0.668, p = 0.002), and BF% (r = −0.890, p < 0.001). Complete autonomic nervous system recovery may depend on the intensity of task demands and cardiorespiratory fitness. Full article

Wildfires have significant ecological, social, and economic impacts, release large amounts of pollutants, and pose a threat to human health. Although deep learning models outperform traditional methods in predicting wildfires, their accuracy drops to about 90% when using remotely sensed data. To effectively monitor and predict fires, this project aims to develop deep learning models capable of processing multivariate remotely sensed global data in real time. This project innovatively uses SimpleGAN, SparseGAN, and CGAN combined with sliding windows for data augmentation. Among these, CGAN demonstrates superior performance. Additionally, for the prediction classification task, U-Net, ConvLSTM, and Attention ConvLSTM are explored, achieving accuracies of 94.53%, 95.85%, and 93.40%, respectively, with ConvLSTM showing the best performance. The study focuses on a region in the Republic of the Congo, where predictions were made and compared with future data. The results showed significant overlap, highlighting the model’s effectiveness. Furthermore, the functionality developed in this study can be extended to medical imaging and other applications involving high-precision remote-sensing images. Full article

In the European Union, Regulation (EU) No 305/2011, in force since 2017 as CPR, requires the classification of cables permanently installed in buildings for reaction to fire, smoke, flaming droplets, and acidity. The latter is an additional classification evaluated through EN 60754-2, involving pH and conductivity measurements. Acidity is the weak point of a PVC cable due to the release of HCl during the combustion. Low-smoke acidity compounds, containing potent acid scavengers at high temperatures, are developed to reduce the acidity of the smoke. In order to design proper HCl scavengers to be used in PVC low-smoke acidity compounds, it becomes essential to evaluate the main actors affecting acidity and conductivity. In this paper, different cable PVC compounds were tested carrying out EN 60754-2 at different temperatures and temperature regimes: measurements of pH and conductivity were compared with ions’ concentration determined by ion chromatography, according to ISO 10304-1 and ISO 14911 for anions and cations, and inductively coupled plasma–optical emission spectrometry, according to ISO 11885. The conclusive results emphasize that HCl from PVC compounds’ thermal decomposition is the primary driver of pH and conductivity, and the contribution from the evaporation and or decomposition of additives and by-products from combustion is found to be negligible in most of the tested PVC compounds for cables. The findings highlight the effectiveness of ion chromatography and inductively coupled plasma–optical emission spectrometry as powerful analytical tools for developing efficient acid scavengers capable of maintaining performance at elevated temperatures. A further outcome regards the experimental demonstration of the limits and incongruencies of EN 60754-2 as an instrument for assessing the additional classification for acidity for cables. Finally, a statistical method to understand through pH and conductivity measurements if the scavenging mechanism acts in the condensed phase is presented. Full article

Fires are classified into five types: A, B, C, D, and F/K, according to the components involved in combustion. Recognizing fire classes is critical, since each kind demands a unique suppression approach. Proper fire classification helps to decrease the risk to both life and property. The fuel type is used to determine the fire class, so that the appropriate extinguishing agent can be selected. This study takes advantage of recent advances in deep learning, employing YOLOv11 variants (YOLO11n, YOLO11s, YOLO11m, YOLO11l, and YOLO11x) to classify fires according to their class, assisting in the selection of the correct fire extinguishers for effective fire control. Moreover, a hybrid model that combines YOLO11n and MobileNetV2 is developed for multi-class classification. The dataset used in this study is a combination of five existing public datasets with additional manually annotated images, to create a new dataset covering the five fire classes, which was then validated by a firefighting specialist. The hybrid model exhibits good performance across all classes, achieving particularly high precision, recall, and F1 scores. Its superior performance is especially reflected in the macro average, where it surpasses both YOLO11n and YOLO11m, making it an effective model for datasets with imbalanced classes, such as fire classes. The YOLO11 variants achieved high performance across all classes. YOLO11s exhibited high precision and recall for Class A and Class F, achieving an F1 score of 0.98 for Class A. YOLO11m also performed well, demonstrating strong results in Class A and No Fire with an F1 score of 0.98%. YOLO11n achieved 97% accuracy and excelled in No Fire, while also delivering good recall for Class A. YOLO11l showed excellent recall in challenging classes like Class F, attaining an F1 score of 0.97. YOLO11x, although slightly lower in overall accuracy of 96%, still maintained strong performance in Class A and No Fire, with F1 scores of 0.97 and 0.98, respectively. A similar study employing MobileNetV2 is compared to the hybrid model, and the results show that the hybrid model achieves higher accuracy. Overall, the results demonstrate the high accuracy of the hybrid model, highlighting the potential of the hybrid models and YOLO11n, YOLO11m, YOLO11s, and YOLO11l models for better classification of fire classes. We also discussed the potential of deep learning models, along with their limitations and challenges, particularly with limited datasets in the context of the classification of fire classes. Full article

A comprehensive understanding of airflow temperature distribution within high-temperature tunnels is crucial for developing effective cooling strategies that ensure a safe environment and acceptable construction costs. In this paper, we introduce a novel cooling strategy that integrates thermal insulation layers and heat exchangers aligned along the tunnel axis (TIL-HE strategy). We investigate variations in airflow temperature and valid ventilation distance (VVD) and compare them with two other cooling strategies: natural tunnels only employing mechanical ventilation (NT strategy) and tunnels featuring thermal insulation layers (TIL strategy), through the 3D k-ε turbulence model in COMSOL Multiphysics. Our findings indicate that (1) the TIL-HE strategy demonstrates superior cooling performance, resulting in significantly lower airflow temperatures and markedly higher VVD; (2) higher water velocity and more heat exchangers contribute to lower airflow temperature and prolonged VVD; (3) positioning the heat exchangers within the surrounding rock rather than inside the insulation layer leads to even lower airflow temperature and longer VVD. Longitudinal-arranged heat exchangers present fewer construction challenges compared to traditional radial-drilled ones, ultimately reducing tunnel construction costs. These findings provide valuable insights for optimizing cooling strategies and engineering parameters in high-temperature tunnel environments. Full article

Globally, fire incidents cause significant social, economic, and environmental destruction, making early detection and rapid response essential for minimizing such devastation. While various traditional machine learning and deep learning techniques have been proposed, their detection performances remain poor, particularly due to low-resolution data and ineffective feature selection methods. Therefore, this study develops a novel framework for accurate fire detection, especially in challenging environments, focusing on two distinct phases: preprocessing and model initializing. In the preprocessing phase, super-resolution is applied to input data using LapSRN to effectively enhance the data quality, aiming to achieve optimal performance. In the subsequent phase, the proposed network utilizes an attention-based deep neural network (DNN) named Xception for detailed feature selection while reducing the computational cost, followed by adaptive spatial attention (ASA) to further enhance the model’s focus on a relevant spatial feature in the training data. Additionally, we contribute a medium-scale custom fire dataset, comprising high-resolution, imbalanced, and visually similar fire/non-fire images. Moreover, this study conducts an extensive experiment by exploring various pretrained DNN networks with attention modules and compares the proposed network with several state-of-the-art techniques using both a custom dataset and a standard benchmark. The experimental results demonstrate that our network achieved optimal performance in terms of precision, recall, F1-score, and accuracy among different competitive techniques, proving its suitability for real-time deployment compared to edge devices. Full article

Ammonia has attracted considerable attention as a zero-carbon fuel for decarbonizing energy-intensive industries. However, its low reactivity and narrow flammability limit efficient ignition and efficient combustion. By using CONVERGR software, this study numerically investigates the ignition and combustion characteristics of liquid ammonia spray ignited by dimethyl ether spray in a constant-volume chamber at an ambient temperature of 900 K. Critical parameters, including injection angles (90°–150°), liquid ammonia injection pressures (60–90 MPa), and ambient pressures (2.8–5.8 MPa), were systematically analyzed to evaluate their effects on ignition conditions and emissions. Results indicate that increasing the injection angle improves mixing between liquid ammonia and dimethyl ether sprays, enhancing combustion efficiency and achieving a maximum efficiency of 92.47% at 120°. Excessively large angles cause incomplete combustion or misfire. Higher liquid ammonia injection pressures improve atomization and promote earlier interactions between the sprays but reduce combustion efficiency, decreasing by approximately 2% as injection pressure increases from 60 MPa to 90 MPa. Higher ambient pressures improve combustion stability but decrease ammonia combustion efficiency. Post-combustion NO emissions at 5.8 MPa are reduced by 60.48% compared to 3.8 MPa. The formation of NO is strongly correlated with the combustion efficiency of liquid ammonia. A higher combustion rate of liquid ammonia tends to result in elevated NO. Based on these findings, an injection angle of 120°, an NH₃ injection pressure of 75 MPa, and an ambient pressure of 3.8 MPa are recommended to optimize combustion efficiency. Full article

The purpose of this study is to understand the law of gas migration in the goaf and reduce the gas on the working face. Taking the N2105 working face of the coal mining industry as the research object, the mathematical model of gas seepage in the goaf was established based on the percolation theory of porous media, and the model was solved. Using Fluent software to simulate the initial pressure, the working face airflow, and gas concentration distribution, different ventilation modes of gas concentration distribution and migration law with different wind speeds after the initial gas pressure. It is concluded that for the first time, the effect of gas on the working face is insignificant, and the influence of the initial pressure on the working surface is gradually revealed. The influence of airflow speed on the goaf is mainly concentrated in the 20~30 m area near the working face, which is affected by the airflow speed of the working face. The gas concentration in the goaf is low, and the fluctuation is obvious. The types of ventilation directly affect the seepage law of goaf gas. The U + I and U + L type ventilation can reduce the gas concentration in the upper corner and f gas seepages from goaf to the working face. Full article

To study the fire behavior of UHVDC (ultra-high-voltage direct current) converter transformers and the effectiveness of CAFs (compressed air foams) in suppressing fires, a full-scale model of a 220 kV converter transformer fire was constructed. The model mainly considered the oil pool fires and oil spill fires that form after explosions, causing the casing to completely fall out. The hot oil fire tests were conducted on the physical converter transformer. The fire suppression characteristics of the CAF system for converter transformer fires were studied. The temperature and changes in various locations of the fire model were analyzed under different foam supply strengths. The fire in a converter transformer is characterized by intense heat, high temperatures, and strong radiation. The highest temperature can exceed 1000 °C in cases of complete combustion. The fire in the converter transformer involves a dynamic oil spill and a large pool of oil, making it challenging to extinguish. The fire extinguishing performance and cooling effect of CAFs are outstanding. The recommended foam supply strength for the actual project is more than 8 L/(min·m²). Full article

The paper addresses the issue of forest fires and critical points in activities related to extinguishing and transport of extinguishing agent to the fire site. With the increasing incidence of forest fires, there are also serious implications for the environment, ecosystems and communities. The relevance of this topic is indisputable, as forest fires are becoming more frequent and intense, with a consequent need for systematic analysis. In this paper, critical sites are identified and assessed, and a description of the equipment used to extinguish a particular fire is provided, with a description of the firefighting strategy in a difficult-to-access site in forest firefighting. This paper shows the effective solution in extinguishing forest fires and then in the design of measures to minimize this risk. We have also assessed the risk activities in this paper. The intent of this article is to show how to effectively extinguish a forest fire. The knowledge gained and recommendations made are aimed at improving firefighter preparedness, techniques and tactics to extinguish forest fires. Full article

Scientometric analysis using the Scopus database and VosViewer program identified the critical directions of development of this or that field to identify promising technologies and to understand how these achievements affect the practice of design and construction. According to the analytics, the average number of articles per year on the topic of structural fire resistance and flame retardants increased by 18% compared to the previous period, and according to preliminary data, the trend will continue in 2024. Among the most cited papers, studies on composite materials and polymers dominate. Among the most productive researchers in the field of flame retardancy of materials are Hu, Yuan (54 papers), Wang, WeiYong (47 papers), and Jiang, Jian (39 papers). According to Scopus, research papers on this topic have been published in 2175 sources. The leading journal in terms of the number of published papers is Fire Safety with 250 publications, but journals such as Fire and Buildings of MDPI Publishing are strongly increasing the pace. Chinese researchers are actively studying various aspects of fire resistance of materials and have published 40% of all papers. Keyword analysis revealed a lack of papers on calculation of fire resistance of structures with fire protection means, calculation of fire resistance of composite structures, and 3D-printed structures compared to the number of articles on the reliability (strength calculation) of building structures. Full article

Tunnel fires often lead to vehicles being trapped inside, causing the “blocking effect”. In this work, fire plume behavior and the maximum ceiling temperature rise in a curved tunnel with blocked vehicles under longitudinal ventilation conditions are studied numerically. The results show that, in curved tunnels, the fire plume in the quasi-stable state exhibits dynamic deflections between the concave and convex walls of the tunnel, so the location of high-temperature zones varies accordingly. The flow field structure in the near field of the blockage and the fire source is complex but can be decoupled into four characteristic sub-structures, i.e., the free shear layer, recirculation I above the vehicle blockage, recirculation II behind the downstream of the blockage, and recirculation III at the top of the tunnel. Recirculation I and II pull the fire plume upstream, while free shear layer and recirculation III pull the flame downstream. The final plume deflection direction depends on the relative strengths of these two pulling forces. As the longitudinal air velocity increases, the plume deflection direction changes from downstream to upstream of the fire source, forming the “downstream tilt—touch the ceiling above the fire source—upstream tilt” mode, resulting in the maximum ceiling temperature rise fluctuating in a decreasing-increasing-decreasing trend. Moreover, the higher the blocking ratio, the lower the critical air velocity required to induce the transition of the plume deflection directions, e.g., a critical wind speed of 3 m/s for a blockage ratio of 0.46 and a critical wind speed of 1 m/s for a blockage ratio of 0.62. Finally, a semi-empirical equation of the maximum ceiling temperature rise in curved tunnels, considering both longitudinal wind and the vehicle blocking ratio, is proposed and validated. This work highlights the multi-dimensional and non-stable plume behavior pattern in a complex tunnel fire scenario, thus providing a deeper understanding to improve the classical tunnel fire dynamic system. Full article

Wildfires significantly impact ecosystems, economies, and biodiversity, particularly in fire-prone regions like the Caatinga biome in Northeastern Brazil. This study integrates machine learning with climate and land use data to model current and future fire dynamics in the Caatinga. Using MaxEnt, fire probability maps were generated based on historical fire scars from Landsat imagery and environmental predictors, including bioclimatic variables and human influences. Future projections under SSP1-2.6 (low-emission) and SSP5-8.5 (high-emission) scenarios were also analyzed. The baseline model achieved an AUC of 0.825, indicating a strong predictive performance. Key drivers of fire risk included the mean temperature of the driest quarter (with an importance of 14.1%) and isothermality (12.5%). Temperature-related factors were more influential than precipitation, which played a secondary role in shaping fire dynamics. Anthropogenic factors, such as proximity to farming and urban areas, also contributed to fire susceptibility. Under the optimistic scenario, low-fire-probability areas expanded to 29.129 Mha, suggesting a reduced fire risk with climate mitigation. However, high-risk zones persisted in the Western Caatinga. The pessimistic scenario projected an alarming expansion of very-high-risk areas to 12.448 Mha, emphasizing the vulnerability of the region under severe climate conditions. These findings underline the importance of temperature dynamics and human activities in shaping fire regimes. Future research should incorporate additional variables, such as vegetation recovery and socio-economic factors, to refine predictions. This study provides critical insights for targeted fire management and land use planning, promoting the sustainable conservation of the Caatinga under changing climatic conditions. Full article