Building Fire Research Articles

Development of a Data-Based Machine Learning Model for Classifying and Predicting Property Damage Caused by Fire

Large fires in factories cause severe human casualties and property damage. Thus, preparing more economical and efficient management strategies for fire prevention can significantly improve fire safety. This study deals with property damage grade prediction by fire based on simplified building information. This paper’s primary objective is to propose and verify a framework for predicting the scale of property damage caused by fire using machine learning (ML). Korean public datasets are collected and preprocessed, and ML algorithms are trained with only 15 input data using building register and fire scenario information. Four models (artificial neural network (ANN), decision tree (DT), k-nearest neighbor (KNN), and random forest (RF)) are used for ML. The RF model is the most suitable for this study, with recall and precision of 74.2% and 73.8%, respectively. Structure, floor, causes, and total floor area are the critical factors that govern the fire size. This study proposes a novel approach by utilizing ML models to accurately and rapidly predict the size of fire damage based on basic building information. By analyzing domestic fire incident data and creating fire scenarios, a similar ML model can be developed.

Building fire safety management analysis based on the reliability assessment method of making safety systems in greenhouse display buildings

Fires must be alerted and anticipated or detected as early as possible. If a fire occurs, it will be fatal such as damage to buildings, infrastructure, other facilities and can even result in casualties. Therefore, a reliable fire protection system must be applied. To prevent or anticipate fires, detection must be carried out as early as possible and prepared facilities or infrastructure for fire safety. The Green House display building is an exhibition building of living plants from various mountains in Indonesia including the mountains of Java, the mountains of Sumatra, the mountains of Kalimantan, and the mountains of Papua. This building is the only building in Indonesia, which presents the atmosphere of the mountains exactly the same as the original mountains, both plants, soil and air temperature, but is in a closed room. Plants and soil are taken directly from the associated location. This building is allegedly very prone to fire because there are so many flammable materials such as leaves, wood and others. As well as forests in Indonesia, especially in Kalimantan and Sumatra where fires often occur. Therefore, the application of Building Fire Safety Management in this building must be applied as well as possible. Fire safety management in question includes: completeness of facilities and infrastructure, active protection systems and passive protection systems. Active protection systems such as fire detection devices for this building cannot use ordinary systems, because in this building, the atmosphere is like mountains with many big trees. To ensure that the building can be declared reliable in terms of building safety, especially fire safety, it is necessary to check the reliability of the building. The inspection can be carried out by several methods, including analyzing based on applicable standards and analyzing based on the assessment of the reliability of the building safety system referring to the Settlement Research and Development Center, Research and Development Agency PU, Department of Public Works in 2005 (Pd-T-11-2005-C).

Creation of Equipment for Testing Roofs for Resistance to External Fire Influence

This paper examines and analyzes the domestic and foreign regulatory framework, guidelines and technical documents, information brochures, manuals, handbooks, test reports and other literary sources as well as information on the Internet. In the countries of the European space, fire safety requirements have been established for roofs and roofing materials, both for roofing materials and for roofs as part of structures, which helps to limit the spread of fire in buildings and structures during a fire. It has been proven that there is a need to create equipment for testing resistance to external fire exposure in accordance with DSTU CEN/TS 1187:2016 (method 2) of coverings of buildings (roofs) and establishing fire classification according to DSTU EN 13501-5:2016. A registration certificate, a technical description, an operating manual, and a program and methodology of the test bench verification have been developed. The initial verification of the test bench was carried out and it was put into operation. Approbation of the test method was carried out on the created test bench through experimental studies of five types of roofing materials being widely used on construction works in Ukraine using two bases (substrates) of combustible and non-combustible materials at air flow speeds of 2 m/s and 4 m/s above the specimen surface. According to the results of this work, it was established that the maximum relative deviation of the result of measuring the length of the damaged (charred) part of the specimens was in the range of 1.2 % to 23.3 % for the roof and 0.7 % to 27.5 % for the base.

Quantification of high temperature stability of mineral wool for fire-safe insulation

Mineral wool, particularly stone wool, is a widely applied thermal insulation material that plays a critical role both in saving energy and in slowing the spread of fire in buildings owing to its high-temperature stability (HTS). However, so far there has been a lack of a universal method to accurately quantify HTS of mineral wool on a small scale. Here, we established a universal method, which is based on measuring the variation of the silhouette area of a cylindrical wool fiber pellet during heating by a hot-stage microscope. Using this method, we detected two main stages of shrinking: 1) the first-stage shrinking related to viscous deformation; 2) the second-stage shrinking caused by melting. Minimizing the first stage shrinking is the key to ensure the fire barrier role of stone wool. The origin of the HTS differences among different types of wool products was clarified by X-ray diffraction, differential scanning calorimetry and thermal expansion.

Evacuation Path Planning Based on the Hybrid Improved Sparrow Search Optimization Algorithm

In the face of fire in buildings, people need to quickly plan their escape routes. Intelligent optimization algorithms can achieve this goal, including the sparrow search algorithm (SSA). Despite the powerful search ability of the SSA, there are still some areas that need improvements. Aiming at the problem that the sparrow search algorithm reduces population diversity and is easy to fall into local optimum when solving the optimal solution of the objective function, a hybrid improved sparrow search algorithm is proposed. First, logistic-tent mapping is used to initialize the population and enhance diversity in the population. Also, an adaptive period factor is introduced into the producer’s update position equation. Then, the Lévy flight is introduced to the position of the participant to improve the optimization ability of the algorithm. Finally, the adaptive disturbance strategy is adopted for excellent individuals to strengthen the ability of the algorithm to jump out of the local optimum in the later stage. In order to prove the improvement of the optimization ability of the improved algorithm, the improved sparrow algorithm is applied to five kinds of maps for evacuation path planning and compared with the simulation results of other intelligent algorithms. The ultimate simulation results show that the optimization algorithm proposed in this paper has better performance in path length, path smoothness, and algorithm convergence, showing better optimization performance.

STUDY OF FIRE LOCATION TIME IN HIGH-STORY BUILDINGS IN CITIES OF UKRAINE

The effectiveness of extinguishing a fire depends on the speed of arrival of fire and rescue units at the scene and the time it will take to deliver the "first" barrel (extinguishing agent) to the location. The rapid supply of fire extinguishing agent gives a chance to localize and then eliminate the fire in the initial phase, although the result depends significantly on the action in the "first minutes of the fire". The urban settlements into appropriate groups based on population and territory using methods of cluster analysis were divided. One settlement was selected from each group of cities, and then data on fires for the period of 2020 was processed. The statistical data obtained from the operational coordination centers of the State Emergency Service of Ukraine in the cities of Kyiv, Kharkiv, Poltava, and Berdyansk, which characterize the process of extinguishing fires, namely the time of arrival of fire and rescue units at the scene of the fire, the time of localization of the fire and the area of the fire, were analyzed. Statistical data were selected only for high-rise buildings. The interrelationship of the following parameters was established: burning area, building floor, time of fire localization. The obtained static data indicate that the area of fire and the time of localization increases on average from 3.5 to 6 times. The result of the data analysis provides a clear understanding of the need to improve the available means and methods (tactics) of extinguishing fires in multi-story buildings. We proposed to use the Typhoon-10 impulse fire extinguisher, which has been converted into a pneumatic gun for the discrete delivery of fire extinguishing substances placed in a special container, as the basis of firefighting tactics in multi-story buildings. Discrete delivery of fire-extinguishing substances in a container allows to reduce the time of localization of fires in multi-story buildings due to the efficiency of deployment of the installation and accurate delivery through the window opening directly to the center of the fire.

Smart fire detection analysis in complex building floorplans powered by GAN

Ceiling-mounted fire service systems are the most widely used provisions to ensure building fire safety. Current design distribution of fire detectors is based on semi-empirical correlations derived from open-floor fire experiments, but building floorplan affects their activation. This work develops a generative adversarial network (GAN) model to achieve accurate and real-time fire detection analysis for buildings with complex floorplans. A numerical fire database with hundreds of floorplans, fire locations and ceiling heights is established to train the GAN model. The pre-trained model can recognize geometric characteristics and reveal hidden fire dynamics laws. Given any building floorplan and fire detector distribution, the model can predict ceiling temperature, velocity and soot density fields with an accuracy of 88% in a second and detection time with 95% accuracy. The proposed GAN model enables a smart fire detection analysis, reduces the fire engineering design cost, and improves fire safety for complex buildings.

Component Modeling for Fire Behavior of Shear Tab Connections

During a building fire scenario, the behavior and capacity of gravity connections can significantly contribute to the integrity of steel-framed building structures. Because gravity connections are subjected to axial and flexural force demands and have a limited rotational capacity due to large beam rotations during a fire scenario, a connection model is needed to simulate their behavior when using analytical models to simulate the behavior of a steel structure in fire. This study develops a component model for shear tab connections at ambient and elevated temperatures in the opensource finite element program, OpenSees, to further enable the use of OpenSees for simulating steel structures in fire. Although the developed component model is not limited to implementation within OpenSees. The developed component model is benchmarked against experimental tests of isolated connections and a structural assembly with shear tab connections subjected to mechanical and thermal loads. Through benchmarking, it is shown that (1) the developed component model could be used to simulate connection behavior during a fire scenario and (2) simulating the ductility of connections and connecting components due to damage is critical when simulating the behavior of shear tab connections exposed to fire.

Assessment of the Performance of FireFOAM in Simulating a Real-Scale Fire Scenario Using High Resolution Data

An assessment of the performance of FireFOAM in simulating a large-scale compartment fire scenario is presented in this study, using the Edinburgh Tall Building Fire Test I (2017) as the basis for evaluation. Different mesh geometries and sizes are tested, and both theory-based and experiment-based validation approaches are employed. The theory-based validation revealed that the implemented finite volumes method is generally conservative, but unaccounted deviations of up to 20% for the heat release rate were observed due to errors in numerically modelling subgrid phenomena, particularly with tetrahedral meshes. In the experiment-based validation, the simulated data showed good agreement with experimental measurements for flow patterns inside the compartment, neutral plane height, and temperatures outside the ceiling jet. However, there were relatively large errors in incident radiation in the hot gas zone, thermal boundary layer transient temperatures, and compartment inflow/outflow velocities. Systematic errors were attributed to deficient heat transfer boundary conditions and under-estimated air entrainment. The study also identified ways to improve run-time efficiency by implementing parallel processing or reducing solid angles in FVDOM, although using large meshes (30 cm and 40 cm cell size) resulted in faster run-times at the cost of accuracy.

The Effect of Plan Corner Shapes on the Spread Speed of Fire in High-Rise Buildings

The Effect of Plan Corner Shapes on the Spread Speed of Fire in High-Rise Buildings

How individuals sense environments during indoor emergency wayfinding: An eye-tracking investigation

Emergency wayfinding is a critical aspect of indoor evacuations. People typically rely on environmental cues to determine their location and navigate their way. However, there is limited exploration into how individuals perceive their surroundings during emergency wayfinding. Through a virtual reality (VR) controlled experiment integrated with eye-tracking, this study explores how individuals sense the environment during evacuations. Participants were immersed in a virtual scenario simulating a building fire and had to choose between Routes 1 and 2 at an intersection for evacuations. The experimental results show individuals' visual attention is positively related to their wayfinding choices. Specifically, familiarity has a significant influence on individuals' visual attention and route choices. Participants who were familiar with Route 2 before evacuations showed increased fixation frequency but smaller saccade velocity during evacuations, reducing their likelihood of detecting the exit sign at the intersection and using Route 1. Neighbor behavior influenced participants' visual information searching and route choices. When virtual neighbors evacuated via Route 1 at the intersection, participants gazed at Route 1 more and were also more likely to use it. The visibility of the exit door on Route 1 also affected both participants' visual attention and route choices. However, the change in emergency lighting did not significantly influence participants’ route choices or their visual attention to Route 1 and the exit sign. These findings enrich our knowledge of human emergency wayfinding and hold significant implications for enhancing building design and evacuation planning.

Simulation Study on Fire Product Movement Law and Evacuation in a University High-Rise Teaching Building

High-rise teaching buildings are complex public buildings that combine the evacuation risks of school buildings and high-rise buildings. In this regard, studying fire product transport patterns and personnel evacuation characteristics of high-rise school buildings is crucial for safe and rapid evacuation. In this paper, we applied Pyrosim2018 software to build a model of a teaching building and performed numerical fire simulation to analyze temperature, CO gas, and visibility to determine the available evacuation time ASET; meanwhile, we performed evacuation simulation by Pathfinder 2019 to determine the required evacuation time and analyze the congestion problem during evacuation. By improving the evacuation route, secondary simulations were conducted and compared with the previous results. The results show that visibility is the main factor affecting evacuation in of the event of a fire in this school building. Based on the visibility analysis, it is recommended that personnel evacuate from floors four and above within 709.2 S when the fire location is at a specific position on the third floor. While the original safety exits of the school building can avoid a large number of casualties, they cannot guarantee the safe evacuation of all people, and planning a reasonable evacuation route can obviously relieve the evacuation pressure in the high-rise corridor.

Localization Strategy of Building Fire Following Earthquake Risk Assessment Method

Localization Strategy of Building Fire Following Earthquake Risk Assessment Method

Experimental study of ejected fire plume from an opening under facing wind passing through the roof

This paper presents an experimental study and analysis of the ejected fire plume from an opening at the top floor of the compartment under facing wind condition. The experiments are conducted by a reduced-scale cubic compartment. An opening facing the wind tunnel exit is set at the center of the compartment side wall. The temperature at the opening and inside the compartment, critical heat release rate for flame ejection, ejected flame morphologic characteristic parameters are measured and quantified by thermocouples and a digital video camera. It's found that the temperature and critical heat release rate for flame ejection first increase then decrease with increasing of facing wind speed. The flame height, flame depth and flame horizontal length increase with heat release rate. The flame depth decreases, flame horizontal length increases as the facing wind speed, while flame height shows a complex evolution with facing wind. A physical model is developed considering mechanisms of flame ejection initial horizontal momentum, flame buoyancy, wind momentum and non-dimensional mass fuel supply rate. The proposed model correlates the experimental data of flame morphologic parameters well. This work considering the practical fire scenario provides an important knowledge supplement on the classical building fire dynamics.

Study on the Location Determination of Building Fire Points Based on Acoustic CT Temperature Measurement

Rapid perception of the location of the fire point is crucial to building fire emergency response in the process of building fire emergency response, which can help firefighters direct fire-fighting operations, effectively control fire sources, and provide strong evidence for the analysis and investigation of fire causes. This paper uses acoustic CT temperature measurement technology to determine the fire source location of a building fire and verifies its validity and applicability as follows: we construct various fire point numerical models based on the fire dynamics simulator (FDS) and obtain temperature data at different times; neural network means were used to obtain the time-of-flight (TOF) of an acoustic wave traveling; the large ill-conditioned matrix equation of acoustic flight under different meshing schemes was constructed and solved based on the simultaneous algebraic reconstruction technique (SART) and least squares QR-decomposition (LSQR), and then reconstruction temperature data under each scheme were obtained. Through the error analysis, the reconstruction effect of each reconstruction scheme is evaluated, and then the applicability of the location coordinate determination of the fire point is analyzed. The results show that the determination of the fire location under the conditions of various fire points in the building space can be realized by acoustic CT temperature measurement technology.

Fire Risk of Polyethylene (PE)-Based Foam Blocks Used as Interior Building Materials and Fire Suppression through a Simple Surface Coating: Analysis of Vulnerability, Propagation, and Flame Retardancy

Building fires can spread through surface combustion of both combustible and interior finishing materials. Recently, the use of foam blocks as interior materials for high-rise residential buildings has increased. However, as foam blocks are primarily composed of polyethylene, they are not flame-retardant and can readily burn and the fire can spread, leading to large-scale damage. Herein, the fire hazard and diffusion characteristics of foam blocks were compared with those of flame-retardant and general wallpapers to confirm the risk of fire. The fire risk of the foam blocks was confirmed using flammability, cone calorimetry, and spread-of-flame analyses. Based on a comparative analysis of the fire risk of foam blocks, the average total heat release was 11.2 MJ/m. This is approximately three times higher than the average heat release rate of the flame-retardant wallpaper and approximately two times higher than that of the general wallpaper. The foam blocks ignited rapidly owing to fire and generated large amounts of combustion gas and heat. To prevent such a fire, 5 wt% montmorillonite (MMT) was simply coated after surface modification to suppress the occurrence of fire. Various flame-retardant materials, surface modifications, and fire safety systems must be developed to prevent fire hazards.

Research on Autonomous and Collaborative Deployment of Massive Mobile Base Stations in High-Rise Building Fire Field.

High-rise building fires pose a serious threat to the lives and property safety of people. The lack of reliable and accurate positioning means is one of the main difficulties faced by rescuers. In the absence of prior knowledge of the high-rise building fire environment, the coverage deployment of mobile base stations is a challenging problem that has not received much attention in the literature. This paper studies the problem of the autonomous optimal deployment of base stations in high-rise building fire environments based on a UAV group. A novel problem formulation is proposed that solves the non-line-of-sight (NLOS) positioning problem in complex and unknown environments. The purpose of this paper is to realize the coverage and deployment of mobile base stations in complex and unknown fire environments. The NLOS positioning problem in the fire field environment is turned into the line-of-sight (LOS) positioning problem through the optimization algorithm. And there are more than three LOS base stations nearby at any point in the fire field. A control law which is formulated in a mathematically precise problem statement is developed that guarantees to meet mobile base stations' deployment goals and to avoid collision. Finally, the positioning accuracy of our method and that of the common method were compared under many different cases. The simulation result showed that the positioning error of a simulated firefighter in the fire field environment was improved from more than 10 m (the positioning error of the traditional method) to less than 1 m.

Preparation and Performance of Building Fire Protection and Insulation Materials

In this paper, a series of calcium silicate thermal insulation materials were prepared by hydrothermal method, and their structure, mechanical properties and thermal insulation properties were systematically studied by adjusting their preparation process, aiming to develop a new type of calcium silicate thermal insulation material with excellent performance. We found that the product gradually changed from a dense flocculent substance on the surface to a loose needle rod substance, and the grain grew, and the crystallinity increased with the increase of synthesis temperature. As the initial calcium silicon ratio increases, the surface pores of the product increase, and the surface gradually becomes loose from dense, resulting in the appearance of tobermorite and xonotlite phases.

Analysis and Improvement of the Fire Incidents in Apartment Houses

In recent years, rapid urbanization and overcrowding have led to the construction of apartment buildings as a means of optimizing land utilization and providing residential facilities. However, among the firefighting facilities installed in these apartments, there is often a lack of automatic sprinklers or fire extinguishing devices in kitchens, particularly in older buildings. This absence of firefighting equipment makes it challenging to extinguish or control fires in their initial stages. To effectively reduce the occurrence of fires in apartment buildings, this study analyzed apartment fire incidents from 2015 to 2021 using the National Fire Data System of the National Fire Service. The analysis revealed that out of a total of 33,807 fires in multi-unit dwellings, 20,236 (59.9%) occurred in apartments. Consequently, there is a need for institutional improvements, as well as increased safety awareness among residents living in apartments and proactive engagement from facility managers.

Development of an Intelligent Detection Method of DC Series Arc Fault in Photovoltaic System Using Multilayer Perceptron and Bi-Directional Long Short-Term Memory

A DC series arc fault is one of the significant sources of electrical wiring fires in residential buildings. The production of extremely high temperatures may lead to the ignition of nearby combustible materials. The applications of arc fault diagnosis based machine learning are a global interest due to the immense challenge to create an accurate and efficient detection method. In this paper, a detection and classification method using a multilayer perceptron incorporated with Bi-Directional Long shortterm Memory (MLP-BiLSTM) is proposed. In order to achieve this goal, nine series arc fault models are used in conjunction with data from real-world observations for simulation purposes using Power System Computer Aided Design (PSCAD) software. The simulation and experimental results confirm that the accuracy of the proposed detection and classification method reaches 99%, which results in that the methodology is believed to be accurate for DC series arc fault detection and classification in the PV system with relatively high accuracy.