• Fire Science and Engineering
• /
• v.33 no.6
• /
• pp.9-19
• /
• 2019

In fire simulations for building fire safety evaluation, changes in the fire area and grid size can significantly influence the prediction results. Therefore, the effects of area changes of the fire source with identical maximum heat release rates on the prediction results of a compartment fire were investigated. The dependence of the prediction results on the grid size using the identical fire area was also examined. No significant changes were observed in the thermal and chemical characteristics of the fires with variable grid sizes, even though the fire area was changed when six or more grids were set based on the fire diameter. In addition, changes in the fire area caused significant differences in the prediction of major physical quantities associated with available safety egress time (ASET) within a compartment. However, the fire area changes did not considerably influence the overall fire characteristics outside the compartment after reaching a certain distance from the opening.

• Fire Science and Engineering
• /
• v.29 no.2
• /
• pp.1-7
• /
• 2015

Tracking, which is one of main reasons of the electric fire, progresses gradually, and therefore, the possibility of fire caused by tracking can be predicted by analyzing the stage of its progress. This paper is conducted in order to predict possibility of the electric fire caused by the tracking in the simulated electric equipment with load. Non-inductive resistance is used as the load. The tracking is happened in a Polyvinyl-chloride-sheathed flat cord, which is a part of the simulated electric equipment by means of dropping of electrolyte droplet. In order to predict the possibility of electric fire caused by tracking, we detect the whole current waveforms of the simulated electric equipment. The time-energy analysis and probability distribution are used for analysis of the tracking progress from the whole current waveforms. In accordance with the results is used for input date of Neural networks, the neural networks can be predict possibility of the electric fire in the electric equipment by 4 stages.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.5
• /
• pp.109-126
• /
• 2023

Early detection of forest fires is essential in preventing large-scale forest fires. Predicting forest fires serves as a vital early detection method, leading to various related studies. However, many previous studies focused solely on climate and geographic factors, overlooking human factors, which significantly contribute to forest fires. This study aims to develop forest fire prediction models that take into account human, weather and geographical factors. This study conducted a comparative analysis of four machine learning models alongside the logistic regression model, using forest fire data from Gangwon-do spanning 2003 to 2020. The results indicate that XG Boost models performed the best (AUC=0.925), closely followed by Random Forest (AUC=0.920), both of which are machine learning techniques. Lastly, the study analyzed the relative importance of various factors through permutation feature importance analysis to derive operational insights. While meteorological factors showed a greater impact compared to human factors, various human factors were also found to be significant.

• Journal of Internet Computing and Services
• /
• v.14 no.4
• /
• pp.85-90
• /
• 2013

Currently, the electric fire prediction system uses PIC(Peripheral Interface Controller) for controller microprocessor. PIC has a slower computing speed than DSP does, so its real-time computing ability is inadequate. So with the basic characteristics waveform during arc generation as the standard reference, the comparison to this reference is used to predict and alarm electric fire from arc. While such alarm can be detected and taken care of from a remote central server, that prediction error rate is high and remote control in mobile environment is not available. In this article, the arc detection of time domain and frequency domain and wavelet-based adaptation algorithm executing the adaptation algorithm in conversion domain were applied to develop an electric fire prediction system loaded with new real-time arc detection algorithm using DSP. Also, remote control was made available through iPhone environment-based app development which enabled remote monitoring for arc's electric signal and power quality, and its utility was verified.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2009.04a
• /
• pp.473-479
• /
• 2009

본 연구는 한국형 도시화재의 물리적 연소성상 예측 모델구축을 위한 기초연구로 일본의 시가지화재시뮬레이션의 구성에 대하여 살펴보고 우리나라의 화재경계지구의 현장조사를 토대로 시뮬레이션을 활용한 Case Study를 수행하였다. 이에 따라 향후 일본과의 국제공동연구의 방향을 설정하여 안전한 방재도시 구현에 이바지하고자 한다.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2009.04a
• /
• pp.270-275
• /
• 2009

최근 소방법의 경우 초고층구조물을 대상으로 한 성능설계 및 화재영향평가등을 실시할 예정으로 있으며 특히 화재위험성평가등에 대한 대책이 요구되고 있으나 이에 대한 데이터가 부족하며 그 방법론 또한 구축되어 있지못한 상황이다. 따라서 본보는 전보에 이어 화재 위험성평가를 위한 방법론에 대한 일환으로서 위험성예측에 사용하는 화재발생의 상위에 의한 화재규모와 rmm 발생율을 각용도별로 기존의 화재에이터 및 가연물조사결과등으로부터 통계적으로 추정하는 방법에 대하여 검토한 것이다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.6
• /
• pp.351-361
• /
• 2021

The seriousness of fire is rising because fire causes enormous damage to property and human life. Therefore, this study aims to predict various risk factors affecting fire by fire type. The predictive analysis of fire factors was carried out targeting Gyeonggi-do, which has the highest number of fires in the country. For the analysis, using machine learning methods SVM (Support Vector Machine), RF (Random Forest), GBRT (Gradient Boosted Regression Tree) the accuracy of each model was presented with a high fit model through MAE (Mean Absolute Error) and RMSE (Root Mean Squared Error), and based on this, predictive analysis of fire factors in Gyeonggi-do was conducted. In addition, using machine learning methods such as SVM (Support Vector Machine), RF (Random Forest), and GBRT (Gradient Boosted Regression Tree), the accuracy of each model was presented with a high-fit model through MAE and RMSE. Predictive analysis of occurrence factors was achieved. Based on this, as a result of comparative analysis of three machine learning methods, the RF method showed a MAE = 1.765 and RMSE = 1.876, as well as the MAE and RMSE verification and test data were very similar with a difference between MAE = 0.046 and RMSE = 0.04 showing the best predictive results. The results of this study are expected to be used as useful data for fire safety management allowing decision makers to identify the sequence of dangers related to the factors affecting the occurrence of fire.

• Fire Science and Engineering
• /
• v.30 no.1
• /
• pp.12-16
• /
• 2016

Propane fires of 1000 to 3000 kW in the ISO 9705 fire room were simulated using FDS to study the problem of decreasing temperature with increasing fire size. A criterion is proposed for under-ventilated fires. The computed temperature at 2000 kW and above was lower than that at 1500 kW. The heat release rate was limited by a lack of oxygen in the simulation. It was found that the heat release rate can therefore be a criterion for under-ventilated fires in simulations. Fires of 1700 kW and above in the ISO 9705 fire room are predicted to be under-ventilated.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2011.11a
• /
• pp.76-79
• /
• 2011

본 연구는 지하대공간의 화재 하중 및 성상을 예측하기 위하여 주요가연물조사를 실시하였다. 조사한 결과 대형서점의 가연물의 종류, 크기, 재질과 지하주차장의 가연물의 종류, 주차대수를 도출할 수 있었다. 이에 대하여 지하공간내 발열량이 높은 곳과 위험성이 높을 것으로 판단되는 곳을 선정하여 가연물조사를 통해 유형을 분석하고, 이에 대한 표준 모델을 모듈화를 제안하여 지하대공간의 화재성상예측을 위한 기초자료로 제시한다.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2011.11a
• /
• pp.163-166
• /
• 2011

본 연구는 Zone Model을 활용하여 장대도로터널의 화재성상 예측방법의 고찰을 목적으로 1/5 Scale 모형실험 및 MLZSUZUKI 결과 분석과 CFD와 MLZ의 비교 분석을 실시하였다. Modeling한 모형터널의 해석시간이 MLZ 1분, FDS 약 6시간으로 CFD에 비하여 360배의 시간 감소와 다층으로 구분하여 구간의 온도변화 및 환기풍에 의한 열기류의 움직임 등의 상황이 해석됨으로 향후 MLZ을 활용하여 터널화재의 위험성 예측에 용이할 것으로 판단된다.