Article
Engineering, Multidisciplinary
Negar Elhami-Khorasani, Juan Gustavo Salado Castillo, Esther Saula, Timothy Josephs, Gauhar Nurlybekova, Thomas Gernay
Summary: This paper discusses a new methodology for conducting fuel load surveys in buildings and applies it to three office buildings. The results show significantly higher fuel load densities compared to older surveys and building codes, particularly in terms of movable items.
Article
Construction & Building Technology
Yongbao Chen, Yunyang Ye, Jingnan Liu, Lixin Zhang, Weilin Li, Soheil Mohtaram
Summary: In this study, three data-driven models (LightGBM, random forest, and LSTM) were developed and compared using a large amount of energy data generated by a physics-based tool. LightGBM outperformed the other two algorithms in terms of prediction accuracy and computation cost.
Article
Construction & Building Technology
Vasco Zeferina, F. Ruth Wood, Rodger Edwards, Wei Tian
Summary: This study examined the sensitivity of cooling demand related results in a large office building using two global sensitivity analysis methods. The findings showed that changes in peak demand were more significant than in annual demand, with ventilation rate being the parameter that contributed the largest for the uncertainty in electricity demand of the HVAC end-use.
ENERGY AND BUILDINGS
(2021)
Article
Construction & Building Technology
Christopher Cole Frishcosy, Yao Wang, Yunping Xi
Summary: The rapid urbanization has led to a more complex relationship between the man-made system and Earth development. Fuel energy, as the main connection between urban and environment, has been a hot spot in recent scientific research. This study proposed a novel method to estimate the mass of fuel load in urban areas, utilizing parameters such as fire load density (FLD) and building types to investigate the distribution of combustible materials. The case study in Washington D.C. demonstrated the significant influence of different land-use types on fuel mass quantities in various districts.
ENERGY AND BUILDINGS
(2021)
Article
Soil Science
Chelene C. Hanes, Mike Wotton, Douglas G. Woolford, David L. Martell, Mike Flannigan
Summary: Forest organic layers are important for predicting fire behavior, emissions, and effects models. Machine learning approaches were used to develop maps of organic layer thickness and fuel load, with random forests being the best method for modeling their spatial distribution.
Article
Environmental Sciences
Victor Fernandez-Garcia, David Beltran-Marcos, Jose Manuel Fernandez-Guisuraga, Elena Marcos, Leonor Calvo
Summary: The study aims to develop generalizable tools for fire danger analysis and predict burn severity. It shows that all environmental variables influence burn severity, and the most effective predictor is the proxy variable obtained through multispectral imagery.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Multidisciplinary Sciences
Rajmund Kuti, Geza Zolyomi, Gabriella Laszlo, Csaba Hajdu, Laszlo Kornyei, Flora Hajdu
Summary: The harmful effects of indoor fires on building structures and the environment are a major concern today. This research aimed to investigate the impact of a typical room fire on the survival chances of trapped individuals and the integrity of the building, specifically focusing on Eastern European architecture. Computational fluid dynamics simulation was conducted to assess the changes in temperature, oxygen levels, and carbon monoxide concentration in a selected room within a vacant building. Based on the simulation results, a 1:1 scale fire experiment was carried out, which confirmed that without the intervention of firefighters, the temperature could rapidly reach 400 degrees Celsius, causing structural damage to the building. Additionally, after 3 minutes, the carbon monoxide concentration reached harmful levels for individuals trapped inside the room, while there was sufficient oxygen at the ground level to support survival for 3 minutes.
Article
Forestry
Cassy Anne Rodrigues, Heloiza Lourenco Zirondi, Alessandra Fidelis
Summary: Fire is a key environmental factor that influences plant communities, with frequency being an important component of fire regimes. Through experiments, it was observed that the quantity, quality, and distribution of fuel impact fire behavior, while fire frequency can alter vegetation structure and affect fire parameters.
FOREST ECOLOGY AND MANAGEMENT
(2021)
Article
Computer Science, Artificial Intelligence
Tiago Pinto, Isabel Praca, Zita Vale, Jose Silva
Summary: This paper presents three ensemble learning models for short term load forecasting, with results showing that the adapted Adaboost model outperforms the reference models for hour-ahead load forecasting.
Article
Construction & Building Technology
Yifei Ding, Yuxin Zhang, Xinyan Huang
Summary: The provision of real-time and detailed evacuation information feedback is crucial for the formulation and adaptation of the onsite evacuation strategy. This study proposes an Intelligent Emergency Digital Twin system that utilizes computer vision and deep learning to detect and track evacuees, calculate their egress speed, and protect their privacy. The system was evaluated in an office building staircase and showed positive results with a high detection rate. The data generated by the digital twin system can be valuable for guiding evacuations in fire scenarios.
JOURNAL OF BUILDING ENGINEERING
(2023)
Article
Chemistry, Multidisciplinary
Sihan Cheng, Zhe Tian, Xia Wu, Jide Niu
Summary: This paper presents a bottom-up high time resolution power load generation method for office buildings, which uses a non-homogeneous Markov chain to establish a random mobility model of personnel and categorizes building electrical appliances into four types based on the different driving modes. By correlating the use of electrical appliances with personnel in the room, a random power simulation model is constructed. Verification results demonstrate the validity and accuracy of the model, as it effectively reflects the daily distribution characteristics of electric load.
APPLIED SCIENCES-BASEL
(2021)
Article
Construction & Building Technology
Hongqiang Fang, Mingjun Xu, Botao Zhang, S. M. Lo
Summary: During building fire emergencies, it is crucial to locate the fire source accurately. This research develops an inverse model for fire source localization using machine learning. By establishing the relationship between fire source location and temperature sensor measurements, the model enables the determination of fire source based on temperature observations. LSTM in stateless mode is found to be more suitable for accurate fire source recognition during the initial and growth stages of fire development.
JOURNAL OF BUILDING ENGINEERING
(2023)
Article
Environmental Studies
Thomas W. McDaniel, Carissa L. Wonkka, Morgan L. Treadwell, Urs P. Kreuter
Summary: A study in Texas and Oklahoma found that officials tasked with implementing burn restrictions have a positive attitude towards prescribed fire as a safe and beneficial land management tool. Familiarity with prescribed fire is a key factor influencing their attitudes, with participation or invitation to participate in prescribed fires playing a significant role. Building trust with local officials and educating fire departments about the benefits of prescribed fire could help reduce burn restrictions and increase opportunities for effective vegetation management.
Article
Energy & Fuels
D. Ramos, P. Faria, A. Morais, Z. Vale
Summary: This paper evaluates the most viable forecasting algorithm for consumption predictions in different contexts and uses decision trees to select the most suitable algorithm. The results show that the decision tree approach can improve the accuracy of predictions.
Review
Ecology
Burhan Rashid Hussein, Owais Ahmed Malik, Wee-Hong Ong, Johan Willem Frederik Slik
Summary: Herbaria contain millions of specimens that have been preserved for scientific studies. Digitization of these specimens is currently ongoing to facilitate easy access and sharing of data. This study reviews the application of computer vision and machine learning techniques to digitized herbarium specimens, categorizes different techniques and applications, and highlights existing challenges and potential solutions.
ECOLOGICAL INFORMATICS
(2022)
Article
Engineering, Civil
Amir Sarreshtehdari, Negar Elhami Khorasani
Summary: This paper investigates the response time of fire departments to ignition locations after an earthquake and proposes a framework that considers both the likelihood of ignition and potential delays in response. This framework incorporates uncertainties in bridge performance and debris accumulation to identify vulnerable areas in a community. A case study is used to demonstrate how the framework can be applied to plan fire department resources and minimize losses from postearthquake fires.
JOURNAL OF EARTHQUAKE ENGINEERING
(2022)
Article
Geosciences, Multidisciplinary
Qi Tong, Thomas Gernay
Summary: This paper proposes the use of hierarchical Bayesian modeling to predict the number of ignitions following earthquakes by analyzing influential variables affecting past ignitions and predicting future ignitions. The model efficiently extracts significant variables and accurately predicts ignition numbers, providing a useful tool for community resilience assessment.
Article
Engineering, Civil
Shuna Ni, Xia Yan, Matthew S. Hoehler, Thomas Gernay
Summary: This paper investigates the thermal and structural response of strap-braced cold-formed steel walls under fire and shear deformation. A numerical model is developed and validated against full-scale experiments to capture the post-fire response and predict the lateral load capacity of the walls.
THIN-WALLED STRUCTURES
(2022)
Article
Engineering, Civil
Carlos Couto, Qi Tong, Thomas Gernay
Summary: This study investigates the application of machine learning models to predict the capacity of steel beams with thin-walled sections at elevated temperatures. Various models such as artificial neural networks, support vector machines, polynomial regression, and random forests are developed, validated, and applied using a large dataset of numerical results. The benefits of using these advanced techniques to calculate the capacity of thin-walled beams at elevated temperatures are demonstrated by comparing them against existing design proposals.
FIRE SAFETY JOURNAL
(2022)
Article
Engineering, Multidisciplinary
Shuna Ni, Thomas Gernay
Summary: Estimating design fires and thermal exposure conditions is crucial in structural fire engineering. This paper summarizes experimental data on the contribution of exposed timber to fire intensity and proposes a method to evaluate design fires and required firefighting resources. The study shows that the surface area of exposed timber is related to the total heat release rate in compartments.
Article
Construction & Building Technology
Yu Xia, Xia Yan, Thomas Gernay, Hannah B. Blum
Summary: A series of new constitutive material models for advanced high-strength cold-formed steels (AHSS) are proposed and fitted to the experimental data, showing accurate description of the stress-strain behavior of AHSS. These results are essential in supporting the adoption of these next-generation steels in future infrastructure.
JOURNAL OF CONSTRUCTIONAL STEEL RESEARCH
(2022)
Article
Engineering, Civil
Ramla Qureshi, Ruben Van Coile, Danny Hopkin, Thomas Gernay, Negar Elhami Khorasani
Summary: Previous research has applied systematic probabilistic models to study the reliability of steel columns under fire. This paper summarizes the established models and applies them to quantify and compare the reliability of steel columns protected based on the US prescriptive approach. The results show that the probability of failure varies significantly depending on factors such as fire protection rating, section factors, and utilization ratio. The thermal properties of insulation have a significant influence on the probability of failure, but limited information is currently available on the behavior of these properties as temperature changes. The influence of fuel load density is presented in the form of fragility functions.
Article
Materials Science, Multidisciplinary
Thomas Gernay, Jochen Zehfuss, Sven Brunkhorst, Fabienne Robert, Patrick Bamonte, Robert McNamee, Siyimane Mohaine, Jean-Marc Franssen
Summary: This paper describes fire tests on loaded glued laminated timber columns, in which the structural response was measured during the heating and cooling phases. The results show that timber columns may fail during the cooling phase after exposure to standard heating.
FIRE AND MATERIALS
(2023)
Article
Construction & Building Technology
Ester Garcia-Castillo, Thomas Gernay, Ignacio Paya-Zaforteza
Summary: This paper compiles available data on the compressive and tensile strengths of timber at elevated temperatures and proposes a probabilistic modeling approach to predict temperature-dependent reduction factors for fire design. The data shows a large scatter in elevated temperature strengths and a conservativeness of the current European standard used for advanced timber design under fire. Multiple probability density functions were calibrated to address this variability, with two-parameter Weibull functions providing the best fit.
JOURNAL OF STRUCTURAL ENGINEERING
(2023)
Article
Engineering, Environmental
Qi Tong, Thomas Gernay
Summary: Facilities in the process industry that deal with hazardous materials are prone to cascading accidents, which can cause extensive damage and performance losses. This study proposes a framework utilizing a dynamic Bayesian network to evaluate the resilience of such facilities. The framework includes quantifiable performance metrics and considers the uncertainties in the evolution of cascading accidents. A case study of a storage tank farm is conducted to demonstrate the application of the framework and evaluate the effectiveness of protection measures through cost-benefit analysis.
PROCESS SAFETY AND ENVIRONMENTAL PROTECTION
(2023)
Article
Engineering, Multidisciplinary
Thomas Gernay, Shuna Ni, David Unobe, Andrea Lucherini, Ranjit Chaudhary, Ruben Van Coile
Summary: In the field of fire safety engineering, cost-benefit analysis is used to determine whether the benefits of implementing fire safety measures justify the costs. However, there is a lack of consistency in the methods used for cost-benefit analysis and a shortage of quantitative data on the costs and economic impact of fire protection in buildings. A recent research project proposes a reference methodology based on Present Net Value evaluation, combined with specialized construction database, fire statistics, and numerical modeling. This paper applies the methodology to four case studies, demonstrating its effectiveness in decision-making when comparing multiple alternatives.
Article
Engineering, Civil
Martin Horacek, Thomas Gernay, Marcela Karmazinova, Marek Hladik, Zbynek Poffel, Ondrej Pesek, Ivan Balazs
Summary: As the use of cold-formed steel profiles for load-bearing structural systems increases, their fire design needs specific attention due to their thin-walled nature. This article describes two standard fire resistance tests on full-scale light gauge steel frame floors made of cold-formed steel, providing new data on the response of unprotected cold-formed steel floor systems subjected to fire.
THIN-WALLED STRUCTURES
(2023)
Article
Engineering, Civil
Thomas Gernay
Summary: This paper provides an overview of the application of the Performance-Based Design approach in structural fire design. The value of the approach is demonstrated through case studies, and the process and role of the structural fire engineer are described. Recent research conducted at Johns Hopkins University, including issues such as the coupling between fire and thermal-structural models, characterization of material behavior at elevated temperature, numerical modeling of structures in fire, probabilistic risk assessment, and cost-benefit analyses, is discussed. The paper concludes with a discussion of challenges and future perspectives for performance-based structural fire design.
FIRE SAFETY JOURNAL
(2024)
Article
Engineering, Industrial
Ruben Van Coile, Andrea Lucherini, Ranjit Kumar Chaudhary, Shuna Ni, David Unobe, Thomas Gernay
Summary: Cost-effectiveness is an important consideration in fire safety engineering, and different approaches exist in the literature for evaluating it. To address this issue, the Fire Protection Research Foundation developed a reference methodology for cost-benefit analysis in fire safety engineering. The methodology sets minimum requirements for assessing cost-effectiveness and emphasizes the use of present net value evaluation for comparing design alternatives. It also suggests that commonly used cost-benefit ratios should only be used when evaluating the effectiveness of a single package of fire safety measures. A case study demonstrates the application of the methodology and highlights the potential sub-optimality of designs based on cost-benefit ratios when evaluating multiple fire safety measures.
Article
Engineering, Civil
Mohsen Zaker Esteghamati, Thomas Gernay, Srishti Banerji
Summary: This study develops explainable data-driven models to predict the fire resistance of timber columns, and compares their predictive capabilities to available prescriptive equations. The results show that the random forest-based model provides the best performance, with accurate and balanced predictions. Column capacity and cross-section dimension are the main factors influencing fire resistance.
ENGINEERING STRUCTURES
(2023)