• Proceedings of the KSR Conference
• /
• 2009.05a
• /
• pp.1984-1988
• /
• 2009

MRT has various surveillance systems for passenger's safety and facility protection which are consisted of fire detection, trespasser observation and so on. However, these systems are not closely related each other because it is designed just for its own purpose, so it could be make wrong decision to surveillance system without important information to determine an accident or disaster. For more accurate event detection, surveillance system needs total situation-aware method using complementary data. This study introduces context-aware system for complex and accurate event detection. Therefore, we apply context-aware system to MRT surveillance system, selecting context-aware parameters and appling them to it.

• Fire Science and Engineering
• /
• v.28 no.6
• /
• pp.52-57
• /
• 2014

The present study suggests the fundamental method for the prediction time of the fire origin by analyzing the combustion phenomenon of inflammable material in the building structure. The heat release rate (HRR) with time variant is evaluated for the interphone as a inflammable material, which is opted from the fire incidents in the stairwell. the fire dynamics simulator (FDS ver. 6.1) is applied in order to analyze the difference of the smoke inflow time to the downstair from the fire event area with various fire pattern. The results show that the maximum inflow time difference for the case of the interphone made from ABS materials is about 4.93 times with the input conditions of heat flux values and the environment in the FDS for the fixed stairwell which composed of total volume $291.3m^3$, floorage $23.3m^2$ and the height of each floor 2.5 m. This research can be practical information for the application method of simulation scheme with experimental data to the fire Identification.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.12
• /
• pp.63-71
• /
• 2020

In this paper, we propose a fire detection technology using YOLOv3 and EfficientDet, the most reliable artificial intelligence detection algorithm recently, an alert service that simultaneously transmits four kinds of notifications: text, web, app and e-mail, and an AWS system that links fire detection and notification service. There are two types of our highly accurate fire detection algorithms; the fire detection model based on YOLOv3, which operates locally, used more than 2000 fire data and learned through data augmentation, and the EfficientDet, which operates in the cloud, has conducted transfer learning on the pretrained model. Four types of notification services were established using AWS service and FCM service; in the case of the web, app, and mail, notifications were received immediately after notification transmission, and in the case of the text messaging system through the base station, the delay time was fast enough within one second. We proved the accuracy of our fire detection technology through fire detection experiments using the fire video, and we also measured the time of fire detection and notification service to check detecting time and notification time. Our AI fire detection and notification service system in this paper is expected to be more accurate and faster than past fire detection systems, which will greatly help secure golden time in the event of fire accidents.

• Journal of Korean Society of Steel Construction
• /
• v.21 no.5
• /
• pp.461-470
• /
• 2009

A column resisting axial load and seismic load is one of the main members in a structural system. The heated column by event of a fire can lose its strength and it may damage its structural system or cause the collapse of the entire structural system. In this study, the fire resistance capacity of internally confined hollow concrete filled tube (ICH CFT) column was investigated. In an ICH CFT column, the yield strength of the external tube is important as a concrete filled tube (CFT) column because the external tube confines the filled concrete and the strength of the column depends on the confined effect. A study was performed by finite element analyses considering the confined effect and material nonlinearity as the temperature changes by the fire. The hollow ratio, the thickness of the external tube, and the strength of concrete were selected as the parameters for the analyses. The analyses were performed by using a commercial FEA program (ABAQUS) and nonlinear concrete model program. The analysis results showed that the hollow ratio and the strength of concrete mainly affect the fire resisting capacity of an ICH CFT column.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.4
• /
• pp.50-58
• /
• 2017

Recently, some methods of converging heterogeneous fire sensor data have been proposed for effective fire detection, but the rule-based methods have low adaptability and accuracy, and the fuzzy inference methods suffer from detection speed and accuracy by lack of consideration for images. In addition, a few image-based deep learning methods were researched, but it was too difficult to rapidly recognize the fire event in absence of cameras or out of scope of a camera in practical situations. In this paper, we propose a novel fire detection system combining a deep learning algorithm based on CNN and fuzzy inference engine based on heterogeneous fire sensor data including temperature, humidity, gas, and smoke density. we show it is possible for the proposed system to rapidly detect fire by utilizing images and to decide fire in a reliable way by utilizing multi-sensor data. Also, we apply distributed computing architecture to fire detection algorithm in order to avoid concentration of computing power on a server and to enhance scalability as a result. Finally, we prove the performance of the system through two experiments by means of NIST's fire dynamics simulator in both cases of an explosively spreading fire and a gradually growing fire.

• Fire Science and Engineering
• /
• v.32 no.4
• /
• pp.25-34
• /
• 2018

This study investigated the development of a fire alarm system with evacuation lighting and voice alarm functions. The performance of a fire detector and system with independently built-in evacuation lighting and voice alarm functions was confirmed for early recognition of fire and to allow visibility of the evacuation route in the event of fire. This new system satisfied model recognition and product testing technological standards with 1.62 lx average illumination, 89.7 dB average sound and 86.1 dB average voice. From additionally testing the evacuation performance of this new system, it was confirmed that the evacuation time decreased by 63.08% to 67.82% under the experimental conditions compared to conventional systems. The new system can minimize fire damage by setting off voice alarms to prevent failure of fire recognition and by flashing emergency lighting to secure the minimum required visibility range for evacuation. Therefore, it is considered that it will be utilized as a fire alarm system with appropriateness and usefulness by considering people with hearing or visual impairment.

• Fire Science and Engineering
• /
• v.31 no.4
• /
• pp.80-85
• /
• 2017

Recently, in Korea, construction of high-rise buildings has been rapidly increasing. Therefore, in order to minimize the loss of life and property in the event of a fire, "performance-based design" which requires performance equal to or better than current regulations is obligatorily required. However, in the field of fire and evacuation simulation, which occupies a large part in the performance-based design, detailed technical guidelines have not yet been established. Therefore, various designers are proceeding with the computer simulation modelling by referring to the design report book previously performed. Especially, in the case of the fire simulation, according to the judgment of a designer the scenario type is selected and the input values is set. Even if the building is used for the same purpose, it is true that the result can be different depending on how and who designed it. Therefore, in this paper, we have investigated the fire scenarios type and scenarios input values by randomly examining 7 preliminary reports of performance-based design in B metropolitan city. We also propose the improvement strategy for fire simulation and lay the groundwork for establishment of technical guidelines for fire simulation for performance-based design.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.6
• /
• pp.1-13
• /
• 2020

In the event of fire, it is necessary to put out the fire within a golden time to minimize personal and property damages. To this end, it is necessary for fire engines to arrive at the site quickly. This study established a fire engine travel time estimation model to secure the golden time by identifying road and environmental factors that influence fire engine travel time in the case of fire by examining data on fire occurrence with GIS DB. The study model for the estimation of fire engine travel time (model 1) covers variables by applying correlation analysis and regression analysis with dummy variables and predicts travel time for different types of places where fire may occur (models 2, 3, 4). Analysis results showed that 17 siginificant independent variables are derived in model 1 and the fire engine travel time differs depending on the types of places where fire occurs. Key variables(travel distance, number of lane, type of road) that are included commonly in the 4 models were identified. Variables identified in this study can be utilized as indicators for research related to travel time of emergency vehicles and contribute to securing the golden time for emergency vehicles.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2017.11a
• /
• pp.56-57
• /
• 2017

Recently, the nation's elderly facilities have grown rapidly since the implementation of the long-term care law in 2008. Also, older adults who use older facilities are also growing. With the increase in the number of fires, increasing the number of fires is also a social issue. In this study, the study conducted an investigation into the smoke control system to analyze the smoke control methods of the elderly and elderly citizens in order to secure the safety of the elderly facilities in the event of a fire, and investigated the smoke control status of the elderly in Korea.

• Wind and Structures
• /
• v.24 no.6
• /
• pp.529-564
• /
• 2017

The Santa Ana winds occur in Southern California during the September-May time frame, bringing low humidities across the area and strong winds at favored locations, which include some mountain gaps and on particular slopes. The exceptionally strong event of late October 2007, which sparked and/or spread numerous fires across the region, is compared to more recent events using a numerical model verified against a very dense, limited-area network (mesonet) that has been recently deployed in San Diego County. The focus is placed on the spatial and temporal structure of the winds within the lowest two kilometers above the ground within the mesonet, along with an attempt to gauge winds and gusts occurring during and after the onset of October 2007's Witch fire, which became one of the largest wildfires in California history.