• Journal of Digital Convergence
• /
• v.10 no.7
• /
• pp.141-152
• /
• 2012

The typical evacuation guidance system based on fire detectors, which is being widely used in theaters and large buildings, is often operated in an analog manner. In case of fire, it often causes the system to lose a wired line or wireless fire detection sensor, resulting in the difficulty of transmitting signals from a wired or wireless fire detection sensor to the main fire monitoring device. Accordingly, this paper has proposed the broadcasting system for disaster management, having an efficient evacuation guidance plan when a disaster occurs. The system reacts to an emergency situation along with fire alarm sirens in real time. We have implemented the above system by means of a simulation program that prints the evacuation guidance information (e.g., location and time of fire, and evacuation path) on an LCD located in a building through the fire sensor network in case of an emergency (e.g., actual fire). We have developed the simulation system by using mathematical algorithms, such as the optimal path search and the fire smoke diffusion algorithm. This simulation program considers the structure of a building and the location where the fire has initially occurred, applying it to the simulator.

• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.1079-1087
• /
• 2018

Recently, computer vision field based deep learning artificial intelligence has become a hot topic among various image analysis boundaries. In this study, flames are detected in fire images using the Faster R-CNN algorithm, which is used to detect objects within the image, among various image recognition algorithms based on deep learning. In order to improve fire detection accuracy through a small amount of data sets in the learning process, we use image augmentation techniques, and learn image augmentation by dividing into 6 types and compare accuracy, precision and detection rate. As a result, the detection rate increases as the type of image augmentation increases. However, as with the general accuracy and detection rate of other object detection models, the false detection rate is also increased from 10% to 30%.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.23 no.6
• /
• pp.52-58
• /
• 2009

When a fire breaks out, it is frequent that large sized miserable death is happened by seriousness of poisonous gas and peculiarity of space because the building construction is recently more complex and diverse. So early countermeasure in preparation for evacuation escape linked directly with a loss of lives is pressing. Because escape light that mark fixing one-way of existing way is not efficiently extricated refugees from dangers when a fire breaks out, construction of system that can extricate refugees from dangers and suppress early a fire by grasping correctly fire point is required urgently. When a fire breaks out, all escape lights connected with fire sensor and reception group which have ill aiming in these point will lead people to safe emergency entrance of opposite direction of place that a fire is broken out after being calculated the direction and speed of flame and smoke. There is the purpose of my research in development of artificial intelligent directional escape light that can mark direction to most suitable pull-out and assist in early extinguishing a fire.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.242-248
• /
• 2019

Fire flame and smoke detection algorithm studies are challenging task in computer vision due to the variety of shapes, rapid spread and colors. The performance of a typical sensor based fire detection system is largely limited by environmental factors (indoor and fire locations). To solve this problem, a deep learning method is applied. Because it extracts the feature of the object using several methods, so that if a similar shape exists in the frame, it can be detected as false postive. This study proposes a new algorithm to reduce false positives by using frame similarity before using deep learning to decrease the false detection rate. Experimental results show that the fire detection performance is maintained and the false positives are reduced by applying the proposed method. It is confirmed that the proposed method has excellent false detection performance.

• Fire Science and Engineering
• /
• v.26 no.1
• /
• pp.102-112
• /
• 2012

In this research, configulation of outside sprinkler system the prevention and postponement of vertical diffusion of blaze was studied prior to this study, vertical diffusion protecting sprinkler head has been developed and the sprinkler system was applied with discharge pressure of 0.05 MPa and flow of 60 l/min witch is stated in NFPA13's Exposure Protection Sprinkler Systems. Through the system design, we applied the system to the sample building and we made pertinent system to work manually and automatically linked to a fire alarm system. Also, we conducted a real-size mock up test verify the cooling effect of the outer wall and the postponement effect of the flame.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.26 no.5
• /
• pp.750-756
• /
• 2002

The effect of equivalence ratio and velocity on the stability of flame in dump combustor was studied in an atmospheric pressure, laboratory scathe dump combustor operating natural gas. Traditionally, peak-to-peak pressure, fluctuation of the heat release rate and Rayleigh index were used to find and control the combustion instability. Cross correlation coefficients, Ci,j which is defined as the normalized value of the integration of the product of two of the mixer pressure, dump plane pressure and heat release rate, are introduced to see whether the flame is stable or not. Ci,j shows more sensitive to combustion status than Rayleigh index in steadily burning flame. Also, the result indicates that the amplitude of Ci,j between heat release and mixer pressure goes up before the flame at the rich de-stabilizing equivalence ratio near $\psi$=0.85. t means Ci:j at this case has a potential to detect the de-stablizing moment in prior to becoming unstable in dump combustor.

• Korean Journal of Materials Research
• /
• v.10 no.11
• /
• pp.778-783
• /
• 2000

Effects of the $Al_2$O$_3$surface protective layer, deposited on the SnO$_2$sensing layer by aerosol flame deposition (AFD) method, on the sensing properties of SnO$_2$thin film ags sensors were investigated.Effects of Pt doping to the $Al_2$O$_3$surface protective layer on the selectivity of CH$_4$ gas were also investigated. 0.3$\mu\textrm{m}$ thick SnO$_2$thin sensing layers on Pt electrodes were prepared by R.F. magnetron sputtering with R.F. power of 50 W, at working pressure of 4mTorr, and at 20$0^{\circ}C$ for 30 min. $Al_2$O$_3$surface protective layers on SnO$_2$layers were prepared by AFD using a diluted aluminum nitrade (Al(NO$_3$).9$H_2O$) solution. The sensitivity of CO gas in the SnO$_2$gas sensor with an $Al_2$O$_3$surface protective layer was significantly decreased. But that of CH$_4$gas remained almost same with pure SnO$_2$gas sensor. This result shows that the selectivity of CH$_4$gas is increased because of the $Al_2$O$_3$surface protective layer. In the case of SnO$_2$gas sensors with Pt-doped $Al_2$O$_3$surface protective layers, low sensing property to CO gas and high sensing property to CH$_4$were observed. This results in the increasing of selectivity of CH$_4$gas selectivity are discussed.

• Fire Science and Engineering
• /
• v.31 no.5
• /
• pp.53-62
• /
• 2017

Electric power which is the energy source of economy and industries requires long distance transportation due to regional difference between its production and consumption, and it is supplied through the multi-loop transmission and distribution system. Prior to its actual use, electric power flows through several transformations by voltage transformers in substations depending on the characteristics of each usage, and a transformer has the structure consisting of the main body, winding wire, insulating oil and bushings. A transformer fire that breaks out in substations entails the primary damage that interrupts the power supply to houses and commercial facilities and causes various safety accidents as well as the secondary economic losses. It is considered that causes of such fire include the leak of insulating oil resulting from the destruction of bottom part of bushings, and the chain reaction of fire due to insulating oil that reaches its ignition point within 1 second. The smoke detector and automatic fire extinguishing system are established in order to minimize fire damage, but a difficulty in securing golden time for extinguishing fire due to delay in the operation of detector and release of gas from the extinguishing system has become a problem. Accordingly, this study was carried out according to needs of active mechanism to prevent the spread of fire and block the leak of insulating oil, in accordance with the importance of securing golden time in extinguishing a fire in its early stage. A bushings fireproof structure was developed by applying the high temperature shape retention materials, which are expanded by flame, and mechanical flame cutoff devices. The bushings fireproof structure was installed on the transformer model produced by applying the actual standards of bushings and flange, and the full scale fire test was carried out. It was confirmed that the bushings fireproof structure operated at accurate position and height within 3 seconds from the flame initiation. It is considered that it could block the spread of flame effectively in the event of actual transformer fire.

• Fire Science and Engineering
• /
• v.32 no.6
• /
• pp.34-39
• /
• 2018

In this paper, the architecture of a safe stove with an automatic fire suppression function using a sound wave fire extinguisher has been proposed and developed for the first time. A microcontroller connected to a fire sensor detects and suppresses a fire by driving a fire extinguisher. The sound wave fire extinguisher is composed of a speaker and collimator, and is driven by a driver module including an audio amplifier. The attenuation of the sound wave is reduced by preventing the sound diffusion with an enclosure surrounding a stove. The frequency of the sound wave is set to 50 Hz, and the sound pressure of 93 dBA is measured at the distance of 0.5 m. It takes maximum 8 and 15 seconds to suppress the flame from 7-cc and 14-cc flammable liquid, respectively, which corresponds to 24% and 42% of the natural extinguishing time. Since the proposed safe stove is non-toxic and leaves no residues over the conventional ones, it would combine with various home appliances to suppress early-stage fires and prevent fire expansion.