• Journal of the Korean Society of Marine Environment & Safety
• /
• v.27 no.1
• /
• pp.22-28
• /
• 2021

This is a basic study on the development of deep learning-based algorithms to detect smoke before the smoke detector operates in the event of a ship fire, analyze and utilize the detected data, and support fire suppression and evacuation activities by predicting the spread of smoke before it spreads to remote areas. Proposed algorithms were reviewed in accordance with the following procedures. As a first step, smoke images obtained through fire simulation were applied to the YOLO (You Only Look Once) model, which is a deep learning-based object detection algorithm. The mean average precision (mAP) of the trained YOLO model was measured to be 98.71%, and smoke was detected at a processing speed of 9 frames per second (FPS). The second step was to estimate the spread of smoke using the coordinates of the boundary box, from which was utilized to extract the smoke geometry from YOLO. This smoke geometry was then applied to the time series prediction algorithm, long short-term memory (LSTM). As a result, smoke spread data obtained from the coordinates of the boundary box between the estimated fire occurrence and 30 s were entered into the LSTM learning model to predict smoke spread data from 31 s to 90 s in the smoke image of a fast fire obtained from fire simulation. The average square root error between the estimated spread of smoke and its predicted value was 2.74.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.7 no.4
• /
• pp.295-304
• /
• 2005

The smoke propagating distances are measured in case that a fire occurs within the subway railroad tunnel. The tunnel is 800m long and the dimension of the cross-section is. Three vertical shafts exist for smoke ventilation. The experiments are performed using the 1/50 reduced-scale model. The smoke propagating distances are measured by thermocouples and by visualization for the accuracy. In order to understand the effect of a fire size and ventilation capacity of the shafts on the smoke propagating distance, 9 test scenarios are chosen. Based on the results, the smoke propagating distance is shown to be important criteria for the ventilation design of the tunnel.

• Clean Technology
• /
• v.14 no.2
• /
• pp.136-143
• /
• 2008

The objective of this study is to investigate the visibility of plume at the Y power plant stack, which fires the orimulsion as a fuel. The plume contains numerous primary particles under $1\;{\mu}m$ size and inorganic ions possibly inferred by the chemicals of secondary aerosol formation. We evaluated the visibility of the plume using the modified PLUVUE-II model. The monitoring data on the particle size distribution (PSD) and secondary aerosols of sulfate were applied to estimate and evaluate the main factors of plume opacity. The chemical reactions were applied to the model for the secondary aerosol formation of $(NH_4)_2SO_4(s)$. The maximum plume length was estimated by an optic method using threshold contrast. The results showed that the plume length was strongly dependent upon the PSD and $(NH_4)_2SO_4(s)$ concentration of the plume emitted from the stack.

• Fire Science and Engineering
• /
• v.21 no.1 s.65
• /
• pp.82-89
• /
• 2007

In this research, we conducted a PIV-visualization experiment to smoke control from fires by using a neighboring Jet fan at minimum distance. We also compared our experiment results with FDS numerical analysis ones. As a result of quantitative data analysis of smoke control modes against fires in tunnel, we concluded that a neighboring Jet fan should be placed at least 25 m away from its fire source in order to produce satisfactory smoke control results. Jet fan showed the best smoke-controlling performance, if placed 50 m away from its fire source. It tended to show an overspeed around the center of tunnel when it is placed more than 75 away from its fire source.

• Fire Science and Engineering
• /
• v.24 no.1
• /
• pp.90-94
• /
• 2010

In this study it is intended to review the moving characteristics of smoke by performing visualization simulation for the calculation of the optimal smoke exhaust air volume in case a fire occurs in tunnels where transverse ventilation is applied, and to obtain basic data necessary for the design of smoke exhaust systems by deriving optimal smoke exhaust operational conditions under various conditions. As a result of this study, if it was assumed 0 critical velocity in the tunnel, the smoke exhaust air volume was limited within 250 meter in the road-tunnel disaster prevention indicator and the exhaust efficiency was from 55.1% to 95.8% in the result of this study. If the wind velocity is in the tunnel, the exhaust rate intends to increase rapidly and the exhaust efficiency is decreased. In addition, if the wind velocity is increased, the exhaust rate should be increased in compared with the generation rate of smoke in maximum 1.8 or 1.04 times. In this study, when the wind velocity is in the tunnel, the limited exhaust rate is $84m^3/s{\cdot}250m$. And if it was assumed 1.75 m/s critical velocity in the tunnel, the exhaust rate would be defined $393m^3/s{\cdot}250m$($Q_E$ = 80 + 5Ar).

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.6 no.2
• /
• pp.129-139
• /
• 2004

The several assumptions and design parameters to determine the ventilation rate in tunnel ventilation system were examined. In longitudinal ventilating tunnel, the ventilation rate has been determined by the critical velocity above which the smoke propagation to the upstream of ventilating air is prevented. Based upon the examination of assumptions and experimental results, we suggested the improved method to determine the critical velocity. In transverse ventilating tunnel, we found that the ventilation rate has been determined in accordance with the custom rather than fire-smoke dynamics such as the critical velocity in the longitudinal ventilating tunnel. It is because the ventilation rate in the transverse ventilation system has been determined by considering only the ventilation of contaminant by vehicle. To improve the ventilation design parameters based upon the fire-smoke dynamics, we conducted model tunnel fire experiments. From the experimental results, smoke propagating distance and smoke filling were suggested as the design parameter to determine the ventilation rate in transverse ventilating tunnel. And tunnels in Europe designed by the custom is found to have the dangerous nature in view of fire safety.

• Fire Science and Engineering
• /
• v.22 no.1
• /
• pp.16-23
• /
• 2008

In this research, the visualization experiment for a scaling tunnel was conducted to establish the optimum fire protection system in tunnel fires. In order to find the optimal operating condition of jet fan with the fire, the characteristics of smoke propagation was considered to find the optimal operating condition of jet fan at the time of tunnel fire, the concentration of smoke was measured experimentally for various jet fan position and it's operating condition. As a result, when jet fan in the vicinity of fire operates at the upstream, the back-layering of the smoke should be considered with separation distance from the fire source. The distance between the jet fan and the fire should be longer than 50 m. On the other hand, when the vicinity jet fan operates at the downstream, the back-layering of smoke does not occur, but stratification is not maintained because the smoke dispersion occurs at the downstream due to the operation of the jet fan.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.499-500
• /
• 2014

화재발생 시 인명 안전을 위하여 초기의 화재감지가 매우 중요한 요인이다. 특히 도로터널, 지하철역사 등 광범위 폐쇄공간에서 연기에 의한 질식사 등 2차 피해의 발생위험이 높다. 이에 최근 광범위 공간에서 적외선 레이저를 활용한 원거리 연기검출 화재탐지기에 대한 연구가 진행되고 있다. 이러한 레이저 기반 원거리 연기검출장치는 이격(100m 이상) 설치되는 레이저 발신기와 수신기의 레이저 포인트가 정확히 정합되어야 한다. 아울러, 레이저 발신기와 수신기 사이의 거리에 비례하여 레이저 초점의 이동거리가 매우 민감하게 변화하므로 이를 정확히 정합하기 위한 고정밀 제어장치가 필요하다. 따라서 본 논문에서는 복수개의 레이저 발신모듈과 복수개의 조도센서(CdS)를 통해 초점의 정합을 자동 추적할 수 있는 알고리즘을 설계, 구현하였다. 이는 초기 레이저 초점의 설정과 이후 외부환경에 따른 초점의 틀어짐을 자동 보정하여 다양한 레이저 인식 장치에 적용될 것으로 사료된다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.497-498
• /
• 2018

In the fire scene, it is difficult to see 1m ahead because of power failure, smoke and toxic gas, even with thermal imaging camera and Xenon searchlight. Analysis of the smoke particles in the fire scene shows that even if the smoke is $5{\mu}m$ or less in wavelength, it is difficult to obtain a front view when using a conventional thermal imaging camera if the visual distance exceeds 1 meter. In the case of black smoke with a particle wavelength of $5{\mu}m$ or more, a space permeation sensor technology using various sensors other than a single sensor is required because chemical materials, gas, and water molecules are mixed. Firefighters need a smoke detection technology for smoke detection and spatial information visualization for forward safety view.In this paper, we design the interface of the information processing system with 32bit CPU core and peripheral circuit. We also implemented and simulated the interface with Lidar sensor. Through this, we provide interface that can implement information processing system of human enhancement fire helmet in the future.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.6
• /
• pp.917-922
• /
• 2013

In this paper, we propose a smoke detection system in order to detect a fire in a wide space, such as tunnel, airports using infra-red and visible laser. The proposed smoke detection system is composed of infra-red laser transmitter and receiver, visible laser and Zigbee wireless communication network. A visible laser is used to match transmitter and receiver and Zigbee network is utilized to propagate warnings of fire. If smoke is appeared between transmitter and receiver, received signals are decreased and it can be considered as occurring smoke. As IR laser transmitter and receiver are separated by long distance, it is difficult to match due to large variations caused by small change of direction. In this paper, it is proposed to match effectively using visible laser. When smoke is detected, warning informations are propagated by Zigbee network in the developed smoke detection system.