• Proceedings of the KSR Conference
• /
• 2007.11a
• /
• pp.1346-1353
• /
• 2007

The simulation analysis of fire-driven flow and passenger evacuation in Daegu subway station, Chung-Ang, have been performed. The first location of outbreak of fire is inside passenger car in the third basement in Chung-Ang station. The smoke flow in the second and third basement has been analyzed using FLUENT 6.2. The CO(carbon monoxide) and temperature distribution in the train units and station platform have been obtained and transferred to input data for evacuation simulation. The highest temperature in the train units was 1500K. For the simulation of passenger evacuation, EXODUS has been used for whole basements (level 1${\sim}$ level 3) in the station. Total number of people was assumed to be one thousand and 640 were placed inside train and 360 were placed outside train. In evacuation simulation, an average of 135 passengers were killed and an average time to evacuate takes 10min 19sec. The main evacuation routes used by passengers were investigated and the cause of death was identified by evacuation simulation.

• Journal of the Korean Society for Railway
• /
• v.11 no.1
• /
• pp.95-100
• /
• 2008

The simulation analysis of fire-driven flow and passenger evacuation in Daegu subway station, Chung-Ang, have been performed. The first location of outbreak of fire is inside passenger car in the third basement in Chung-Aug station, The smoke flow in the second and third basement has been analyzed using FLUENT 6.2. The CO (carbon monoxide) and temperature distribution in the train units and station platform have been obtained and transferred to input data for evacuation simulation. The highest temperature in the train units was 1500k. For the simulation of passenger evacuation, EXODUS has been used for whole basements (level 1${\sim}$level 3) in the station. Total number of people was assumed to be one thousand and 640 were placed inside train and 360 were placed outside train. In evacuation simulation, an average of 135 passengers were killed and an average time to evacuate takes 10min 19sec. The main evacuation routes used by passengers were investigated and the cause of death was identified by evacuation simulation.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2007.04a
• /
• pp.355-360
• /
• 2007

• 한국전산유체공학회:학술대회논문집
• /
• 2010.05a
• /
• pp.501-504
• /
• 2010

In this study, fire behavior for subway coach fire in the 1/10 reduced scale model is analyzed using numerical analysis method (FDS). The size of the reduced model coach is 1.9 m long, 0.290 m high and 0.235 m. The simulation is carried out only one-sided four doors of the coach are opened. A fire source is the n-heptane pool fire with a mean heat release rate 2.33 kW. Smoke temperature, heat release rate and mass loss rate for the model are calculated.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2012.04a
• /
• pp.85-88
• /
• 2012

본 연구는 지하철 객실 화재시 배출가스나 연기에 의한 사망사고에 대해서 FDS(ver. 5.0) 사용코드를 이용해 객실내 화재의 유동현상을 검토한다. 시뮬레이션 모델로 인천메트로 송도연장선을 기준 모델로 설정하였으며 객차 화재발생시 측창문의 개방개소에 따른 배연능력을 검증하기 위하여 측창문 개방조건에 따른 배연특성을 시뮬레이션 하여 인체의 유해성 여부에 따른 위험성 평가에 대한 결과를 도출하였다. 본 연구결과는 지하철의 성능위주 설계를 적용한 기초자료로 활용됨으로써 지하공간 화재시 인명피해 최소화를 목표로 한다.

• Proceedings of the KSR Conference
• /
• 2006.05b
• /
• pp.214-220
• /
• 2006

• Journal of the Korean Society for Railway
• /
• v.7 no.3
• /
• pp.180-185
• /
• 2004

The present study investigates the effect of fire growth model on fire development characteristics in a carriage. The parallel processing version of FDS code is used to simulate the fire driven flow in a carriage and two types of fire growth model which are flame spread model and t$^2$ model are examined for the same geometrical condition. The heat release rates(HRR) of both model are similar each other until 30 s after ignition, but the flame spread model predicts 5 times higher than those of the t$^2$ fire model during the quasi-steady fire period. Maximum heat release rate in the case of flame spread model reaches about to 12 MW at 100 s after fire ignition. Also, various database of fire properties for combustible materials and more elaborate combustion model considering the flame spreading phenomena are required for better predictions of fire development characteristics using numerical simulation.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2006.04a
• /
• pp.128-133
• /
• 2006

• Fire Science and Engineering
• /
• v.22 no.5
• /
• pp.1-8
• /
• 2008

The study is to perform numerical analysis of train fire characteristics in an underground subway tunnel, depending the different locations of ventilation facility. To study the characteristics of train fire, two kinds of worst-case scenarios are selected, based on escape distance, escape time, and fire zone, and trends and thermal environments of tunnel are analyzed by changing the locations of ventilation facility for times. Fire characteristics is calculated by using FLUENT v.6.3.26, and turbulent flow is calculated by using the standard k-${\varepsilon}$ model. The numerical results show distribution of carbon monoxide concentration, temperature, and velocity. The results of this study will contribute to building the most suitable ventilation systems when designing subway stations and tunnels.

• Journal of the Korean Society for Railway
• /
• v.9 no.3 s.34
• /
• pp.291-297
• /
• 2006

The present study investigated the effect of fire suppression using a mid-low pressure water mist in a carriage fire. The fire extinguishing time and temperature distributions below ceiling in the enclosed compartment of $2.9m{\times}2.8m{\times}5.0m$ were measured by stopwatch and k-type thermocouples for various fire positions. The numerical simulations were extensively performed using. Fire Dynamics Simulator(FDS, Ver. 4.0) code and the predictions were compared with experimental data. The prediction results showed good agreement with the measured maximum temperature in the all cases. Whereas the predicted temperature was about $40^{\circ}C$ higher than the measured one after operating of water mist. The predicted fire extinguishing times were compared with those of measured data. Fires are extinguished within 200 seconds at the experiment in Case 2 and Case 3. But in Case 1 fire was not extinguished in the numerical simulation. The reason of the discrepancy between predicted and measured data was that a simple suppression algorithm has been implemented in FDS. Also, various databases of fire properties for combustible materials and more elaborate model considering the water mist were required fur better predictions of the cooling and suffocation effect.