Fire Technology最新文献

Aqueous film forming foam (AFFF) has been the industry standard for combatting liquid fuel fires and hazards for almost 50 years. AFFF is a water-based solution that contains a fluorinated, film forming surfactant [per- and poly-fluoroalkyl substances (PFAS)] to seal the fuel surface during suppression/extinguishment. All “AFFFs” contain PFAS. Many PFAS are classified as forever chemicals (e.g., the ones used in AFFF) that do not naturally breakdown in the environment and/or in the human body. Some PFAS have been associated with health effects in both humans and in some animals. As a result, the ability to use AFFF to extinguish liquid fuel fires continues to be greatly restricted and has already been banned in numerous States in the United States and in countries across the world such as Australia. This article provides an update of the status of AFFF, the available alternatives and any revisions to applicable codes and standards.

Experimental analyses and numerical simulations are carried out on a test case involving an heptane pool fire within a large under-ventilated environment. During the experiments, the temperature history at several locations within the room is monitored by means of thermocouples, and the fire radiative heat transfer estimated through a plate thermocouple. The experimental layout is then replicated numerically and tested using OpenFOAM CFD code. The study is a preliminary analysis performed for code validation purposes on a full-scale fire scenario. The results of the simulations are compared to the experimental results and critically analysed, finding a reasonable agreement overall. Critical issues in fire modelling are also highlighted. In fact, due to the problem complexity and the limitations of the numerical models available some important aspect that can significantly influence the outcome of the simulations must be calibrated a posteriori, somewhat limiting the general predictive applicability of the fire models. Primarily, these are the heat release rate history, the combustion efficiency, and, to a lesser extent, the convective heat transfer boundary condition at the wall.

The mechanical properties and nonlinear performance of the Q960 cold-formed thick-walled ultra-high-strength steel (CTUS) after elevated temperatures were investigated experimentally, where the effects of cold-forming process, elevated temperature, and cooling condition were considered. Seven different elevated temperatures and two different cooling conditions were included in experiment, where a total of 45 coupon specimens were tested. The tensile coupon test was performed on the Q960 CTUS specimens to reveal the influences of the elevated temperature and cooling condition on the residual stress–strain properties. The predictive equations for the key mechanical parameters of the Q960 CTUS after elevated temperatures were proposed based on the experimental results. The mechanical properties of the Q960 CTUS after elevated temperatures were compared with those of different structural steels and reinforcing steels. A reliability analysis was performed to determine the accuracy of predictive equations for key mechanical parameters, where the resistance factor was recommended. A constitutive model was suggested to elucidate stress–strain curves of the Q960 CTUS after elevated temperatures. These research findings served as the foundation for the future numerical and theoretical investigations on the residual resistant performance of the Q960 CTUS structures after fire.

The goal of the present work is to establish a framework for firebrand morphology characterization. Central to this framework is the development of a simple firebrand shape classification model using multi-dimensional particle shape descriptors. This classification model is built upon a series of synthetically generated 3D particles whose shapes and sizes are chosen to be representative of actual firebrands typically encountered during vegetative and structural fuel burns. Principal Component Analysis (PCA) is applied to the synthetic dataset and used to structure the classification model. The model is then verified using 3D digital representations of real-world particles (firebrands collected during tree burns and unburned bark pieces from oak trees). The classification model, which will allow meaningful comparisons of firebrand morphological features by shape class, is expected to be gradually refined as more datasets are made available throughout the Wildland–Urban Interface (WUI) fire research community.

This study aims at utilizing machine learning (ML) in predicting the fire resistance and spalling degree of reinforced concrete (RC) columns with improved accuracy and reliability. A database with 119 test specimens was created for the development of ML-based regression models, and a database with 101 test specimens was created for the development of ML-based classification models. Six ML algorithms—support vector machine (SVM), random forest (RF), multilayer perceptron (MLP), extreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), and light gradient boosting machine (LightGBM). The hyperparameters of the ML-based models were optimized through Bayes optimization search (BayesSearchCV) with ten-fold cross-validation. The results indicated that the AdaBoost not only accurately predicted the spalling degree of RC columns with an accuracy of 87%, but also performed best in predicting the fire resistance of RC columns with R² = 0.96 and RMSE = 16.58. The AdaBoost model achieved high accuracy without significant bias, surpassing existing design equations. SHAP method was utilized to produce global explanations for the predictions. The results revealed that concrete compressive strength, loading ratio, slenderness ratio, and column width were the most critical features for spalling degree identification. Meanwhile, those were slenderness ratio, concrete cover, loading ratio, part of the fired column, and longitudinal reinforcement for fire resistance prediction. The parametric study demonstrated that the fire resistance of RC columns is positively affected by only concrete cover.

This work concerns how to effectively detect the fire in early stage using computer vision method. As known, the flame of early fire is small and cannot be effectively detected by traditional fire detectors. Inspired by color characteristics of flame, we proposed a Shallow Guide Deep Network (SGDNet) to address the problems in existing early fire detection models. We first investigate the feature of fire in YCbCr color space, then design an SGD module to fuse shallow features, so as to guide the fusion of deep features. Backbone, anchors, head and IoU of model are redesigned according to the features of early fire to not only fuse the deep features but also reduce the size and infer time. Finally, we implement a Early Stage Fire Detection System based on our SGDNet, using embedded device as computing platform, connecting 4 IP cameras for test. Multithread is widely utilized in system for detecting and the reading and conversion operations of video streams, which effectively improves the execution efficiency and reduces the delay of system. Experimental results on dataset show high performance of our model with the advantage of small size and parameter. Application in actual scenarios proves that the delay for detection is about 1.2 s, which fulfills the requirement of early fire warning.

Fire accidents are more likely to occur in tunnels with different curves, aspect ratios, and slopes due to the land’s geographical characteristics. A three-dimensional computational fluid dynamics code with curvilinear grids fitted to the body was used to simulate a variety of fire locations releasing heat at a rate of 5 MW–60 MW in a tunnel with a turning radius of 100 m–1500 m, an aspect ratio of 0.5–2, and a slope between – 10% and 10%. Using the Design of Experiments (DOE) method coupled with numerical simulations, 32 3D numerical models were constructed and a second-order critical velocity model was generated. Analysis of critical velocity was performed based on Response Surface Methodology (RSM) and multifactor curve plots were drawn for effective parameters. The results showed that the critical velocity was proportional to one-third power of the heat release rate. It was also observed that the critical velocity increased gradually as the fire source moved from the tunnel’s center to its walls. Furthermore, the critical velocity decreased with increasing the aspect ratio. Results showed that the critical velocity increased with increasing the tunnel turning radius. Moreover, tunnels with negative slopes have a higher critical velocity than horizontal tunnels without slopes. Finally, by defining the parameters in non-dimensional form, a new modified form was derived for critical velocity calculation (R² = 0.98). A critical velocity of 1.24 m/s–5.21 m/s was calculated based on five parameter values in this study. Furthermore, other straight and curved tunnel models confirmed the formula’s predictions.

The traditional storage method of fire accident cases is mainly in the form of text, and it is difficult to effectively conduct comprehensive analysis due to the limited ability to display key information and fire knowledge. In this paper, a structured storage form of building fire cases was proposed based on knowledge graph, which can comprehensively describe and visualize the fire causes, the dynamic fire development process and evacuation process. It enables readers to get information and knowledge from building fire cases intuitively, and supports the comprehensive analysis for building fire prevention strategies. The knowledge graphs are constructed for two common building types (residential and public buildings), and have the capacity to reflect the dynamic development law of fires from ignition to spread in different buildings. Meanwhile, as the occupants’ evacuation is the first concern when a fire occurs, the knowledge graphs also visualize the relationship among various conditions in the evacuation process. Different application scenarios are displayed in the paper, including case query, root-cause analysis and consequence forecasting, which shows the advantages and applicability of building fire knowledge graph.

The mechanical material behavior of mild steels is reversible in the cooling phase of natural fires, which is proven by experimental evidence. For the material behavior of high-strength steels during cooling, no results are yet available. The paper provides the first comprehensive test program on the constitutive material behavior of high-strength steels S690QL and S960QL as well as mild steel S355 J2 + N in the case of natural fires. It is elaborated that the mechanical material behavior of high-strength steels in the cooling phase differs from the behavior in the heating phase and is not reversible due to phase changes of the microstructure. A constitutive material model for structural fire design purposes is developed on the basis of experimental data and the soundness and reliability of the model are proven by a statistical study that systematically evaluates the deviation of the model prediction from the test data.