• 보존과학연구
• /
• s.31
• /
• pp.155-171
• /
• 2010

Wooden cultural heritages have many factors of fires and structural characteristics vulnerable to the fire extinguishing. Also, they are surrounded with forests and so remote from fire stations, which make it difficult to handle it quickly when fires break out. Wooden cultural heritages made of wood materials belong to the general fire in a Class A. Taking characteristics such as a smoldering and a backfire from the that fire of wooden materials into consideration, extinguishing the fire by the cooling system is the most effective. If the fire can't be put out at the early stage, it is almost impossible to protect wooden cultural heritages from the fire, because wooden structures can be destroyed in a high temperature and in a short time and it takes around average 7 minutes to reach its peak of flames in the process of a fire. According to the geographical and environmental situation of the cultural heritages, currently, the fire-prevention facilities such as the auto fire detector for the prompt detection, the water mist fire suppression system for the 1st early and urgent fire suppression and the outdoor fire hydrant and the water curtain etc. for the 2nd full-scale suppression and the prevention of the fire gaining force are being installed for the wooden cultural heritages.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.12
• /
• pp.320-330
• /
• 2020

Fire is difficult to achieve good performance in image detection using deep learning because of its high irregularity. In particular, there is little data on fire detection in underground utility facilities, which have poor light conditions and many objects similar to fire. These make fire detection challenging and cause low performance of deep learning models. Therefore, this study proposed a fire detection model using deep learning and estimated the performance of the model. The proposed model was designed using a combination of a basic convolutional neural network, Inception block of GoogleNet, and Skip connection of ResNet to optimize the deep learning model for fire detection under underground utility facilities. In addition, a training technique for the model was proposed. To examine the effectiveness of the method, the trained model was applied to fire images, which included fire and non-fire (which can be misunderstood as a fire) objects under the underground facilities or similar conditions, and results were analyzed. Metrics, such as precision and recall from deep learning models of other studies, were compared with those of the proposed model to estimate the model performance qualitatively. The results showed that the proposed model has high precision and recall for fire detection under low light intensity and both low erroneous and missing detection capabilities for things similar to fire.

• International Journal of Safety
• /
• v.11 no.1
• /
• pp.15-18
• /
• 2012

Because of development of printing technology, toner cartridge particle becomes smaller and more dangerous. And sometimes we had incidents with dust explosion of toner cartridge particle at recycling facilities in Japan. Therefore we studied on hazard of toner particle relating with dust explosion. We found that toner particle is so dangerous compared with most organic solids, even though it does not belong to hazardous materials in the UN regulation and the Japanese Fire Service Law.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2007.04a
• /
• pp.95-100
• /
• 2007

• Journal of the Korea Safety Management & Science
• /
• v.15 no.4
• /
• pp.105-111
• /
• 2013

This study aimed to identify the recognition of fire risks on their field operations among firefighters and contribute to setting up safety measures accordingly. A self-administered questionnaire were used to evaluate their recognition of job risks. About 93.4% of respondents answered they have exposed to job risks and 73.4% answered that they had high safety awareness. Our study showed that the lower service years and ranks the fire officers had, the lower relevance in statistics their awareness of the risk of the field operations and safety concerns had. The vulnerable group is those aged 20 to 39, with under 10 years of service, at a lower rank of senior fire sergeant. The study suggests coming up with the measures to motivate the officers to perform safe operations by improving their safety awareness.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.15 no.6
• /
• pp.658-663
• /
• 2015

In the previous work, we have proposed an integrate-and-fire neuron circuit and synaptic device based on the floating body MOSFET [1-3]. Integrate-and-Fire(I&F) neuron circuit emulates the biological neuron characteristics such as integration, threshold triggering, output generation, refractory period using floating body MOSFET. The synaptic device has short-term and long-term memory in a single silicon device. In this paper, we connect the neuron circuit and the synaptic device using current mirror circuit for summation of post synaptic pulses. We emulate spike-timing-dependent-plasticity (STDP) characteristics of the synapse using feedback voltage without controller or clock. Using memory device in the logic circuit, we can emulate biological synapse and neuron with a small number of devices.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.25 no.5
• /
• pp.255-260
• /
• 2013

Since a chilled beam system can reduce the conveyance energy for air conditioning and has many advantages in terms of constructability and indoor environment, its supply has been rapidly expanding mainly in Europe. For this chilled beam system to be effectively available in Korea, a product suitable for the domestic climatic environment and HVAC system operating conditions needs to be developed. Also, domestic products need to have an edge in cost competitiveness compared to overseas products. In this study, the induction structure of a chilled beam is deduced and an analysis on the induction performance is carried out while the chilled beam system that meets domestic situations is in the process of development.

• International Journal of Air-Conditioning and Refrigeration
• /
• v.12 no.1
• /
• pp.30-36
• /
• 2004

The present study investigates the fire suppression characteristics using a water mist fire suppression system. Numerical simulations of fire suppression with water mist are performed with considering the interaction of fire plume and water spray. The predicted temperature fields of smoke layer are compared with those of measured data. Numerical results agree with the experimental results within $10^{\circ}C$ in the case without water mist. In the case of fire suppression with water mist, numerical results do not predict well for temperature field in the gradual cooling region after water mist injection. But the predicted results of initial fire suppression are in good agreement with those of measured data. The reason for the discrepancy between predicted and measured data is due to the poor combustion modeling during the injection of water mist. More elaborate models for numerical simulation are required for better predictions of the fire suppression characteristics using water mist.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2019.05a
• /
• pp.4-5
• /
• 2019

Smoke and fire spread throughout the building as a result of problems in the fire prevention section of Uijeongbu Apartment Fire in 2015 and Jecheon Sports Center Fire in 2017 in Korea, causing a number of casualties. It is believed that the fire compartment standards need to be reviewed as the fire in recent buildings continues to show problems. In this study, a comparative analysis was conducted on the fire prevention zone standard with Japan to improve the fire compartmente standard in Korea.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.16 no.3
• /
• pp.83-95
• /
• 2013

To understand natural regeneration and stand development after fire in mixed broadleaved-coniferous forests of Sikhote-Alin Mountains, ten sample plots of $50m{\times}50m$ size were established in 1975 and 1983 at the stands burned by wildfires in 1973 and 1982, respectively. And, the number of naturally regenerated seedlings were monitored in two $50m{\times}4m$ subplots in each plot. The most fire-sensitive conifer species is Abies nephrolepis, while Betula costata is the most fire-sensitive broadleaved tree species. The most fire-resistant species were Q. mongolica, T. taquetii and A. mono. The results of 20 and 30 years after the fire showed that pioneer tree species, e.g. Populus, Salix, and Betula, were regenerated immediately at the early stage of stand development and grew where there is a mono canopy layer with high density. On the other hand, the densities of successors, e.g. Pinus koraiensis, Picea jezoensis, Abies nephrolepis, Acer mono and Tilia taquetii, which were present in the study plots before the fire, increased gradually. Naturally regenerated tree species after forest fire by the growth rate were divided into three groups according to their annual height growth. The seral tree species (Betula costata, Betula platyphylla, Padus maackii, Populus tremula and Sarix caprea) belong to the first group and have the highest growth rate (from 40 to 96 cm per year). The late successional broad-leaved trees (Tilia taquetii, Acer mono and Quercus mongolica) belong to the second group and have intermediate annual height growth (from 3.7 to 13.5 cm per year). The late successional coniferous species (Picea jezoensis, Pinus koraiensis and Abies nephrolepis) form the third group and have the least annual height growth (from 1.4 to 3.5 cm per year).