• Journal of ILASS-Korea
• /
• v.25 no.3
• /
• pp.119-125
• /
• 2020

In this paper, we used Helmholtz solver based on 3D finite element method to quantitatively analyze the effects of change of gain, time delay and time delay spread, which are the main variables of flame transfer function, on combustion instability in gas turbine combustor. The effects of the variable of flame transfer function on the frequency and growth rate, which are the main results of combustion instability, were analyzed by applying the conventional heat release fluctuation model and modified one considering the time spread. The analysis results showed that the change of gain and time delay in the same resonance mode affected the frequency of the given resonance modes as well as growth rate of the feedback instability, however, the effect of time delay spread was not relatively remarkable, compared with the dominant effect of time delay.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.21 no.1
• /
• pp.125-131
• /
• 1997

The extinction of premixed flames under the influence of stretch is studied numerically. A wide range of fuel (hydrogen, ethylene, acetylene, methane, propane and methanol) and air mixtures are established in an opposed jet and their flame properties such as flame speed, flame thickness, thermal diffusivity, and stretch rate at extinction are computed. Computations are made using several chemical kinetic mechanism (Smooke, Kee et al. and Peters). The major result is that, in contrast to the various previous claims of extinction Karlovitz number varying over three orders of magnitude, it is found to be constant around two for all of the mixtures tested. That is, premixed flames are extinguished when the physical flow time decreases (due to increased stretch rate) to the point where it approximately equals the chemical reaction time. Here the relevant chemical reaction time is not the one computed using the one-dimensional flame properties as originally suggested in the formulation of Karlovitz number, but rather it is the one obtained using the stretched flame properties which fully reflect the effect of straining on the flame structure.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2023.11a
• /
• pp.29-30
• /
• 2023

Wood is a construction material that has the advantages of carbon dioxide storage ability, noise reflection, and eco-friendliness. In order to use wood for a long time, you must use wood-specific paint, which is called oil stain. Oil stain improves water resistance and moisture resistance, but has the disadvantage of being weak against fire. This is because the oil contained in the oil stain causes a chemical reaction, and this chemical reaction causes the oil stain to spontaneously ignite, igniting nearby combustible materials and causing frequent fires. To improve this, in this study, different flame retardants were mixed and added to oil stain to produce functional oil stain. In addition, we would like to apply it to wood to check glow time and carbonization area. As a result of the experiment, it shows the best performance when mixed at 30(15 + 15)(%) and added to oil stain. The remaining burn time is satisfied from 10% for all samples, and the carbonized area is satisfied when it is 30%.

• Journal of the Korean Society of Safety
• /
• v.13 no.3
• /
• pp.119-125
• /
• 1998

In long vertical duct, the aspect of second flame in laminar flame propagating through lycopodium-air mixtures and the behavior of dust particles in neighborhood in front of flame have been examined experimentally. In order to trace the development of second flame to its origin, the velocity and vorticity distribution of dust particles in front of flame were measured by using with the real-time PIV system. The velocity of particles was approximately zero at the central part of flame front and the ahead of the flame leading edge, but maximum near the duct wall. The flame velocity of second flame and the movement of leading flame edge depend mainly on behavior of dust particles by the flow distribution of temperature and pressure.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.20 no.3
• /
• pp.1040-1049
• /
• 1996

The high loading combustion is accomplished by making the turbulent intensity strong and the scale small in the premixed combustor. The Da-mkoler number, which is decreased by short turbulent characteristic time or by long chemical reaction time, can make the distributed reaction flame. So we developed a doubled jet burner for high loading combustion. The doubled jet burner was designed to make the scale of the flame small by the effect of impingement and increasing shear stress with doubled jet. We investigated the turbulence characteristics of unburned mixture and visualized several flames with the typical schlieren photography. Then we studied the influence of several factors that related the scale of flame. Consequently, the doubled jet burner can make the eddy very small. And we can obtain the detail information of the flame scale through ADSF(the Average Distance between Successive Fringes) in the micro- schlieren photography. The ADSF is not a exact flame scale, but it has qualitative trend with increasing turbulent intensity. The ADSF is diminished remarkably with increasing turbulent intensity. The reason is that strong turbulent intensity makes the flame zone thick and flamelets numerous. We can confirm this fact by the signal analysis of ion currents.

• Journal of the Korean Society of Combustion
• /
• v.8 no.2
• /
• pp.50-55
• /
• 2003

Synthesis of carbon nanofibers on a metal substrate by an ethylene fueled inverse diffusion flame was observed. Stainless steel plates were used for the catalytic metal substrate. The effects of radial distance and residence time of the substrate were investigated. The role of hydrocarbon composition in the fuel was also viewed. Nanofibers with a diameter range of 30-70nm were found on the substrate. The carbon nanofibers were formed and grown in the region from 4 to 5.5mm from the central axis of a flame outside of the visible flame front in the radial direction. The minimum residence time required for the formation of carbon nanofibers were about 20 seconds, and over 60 seconds were required for the full-scale growth. The characteristic time of the formation of carbon nanofibers was much shorter than that of the substrate temperature growth. In this study, the variation in hydrocarbon composition had no significant effect on the formation and growth of the carbon nanofibers.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.112-119
• /
• 2019

In this paper, we propose a deep learning structure suitable for embedded system. The flame detection process of the proposed deep learning structure consists of four steps : flame area detection using flame color model, flame image classification using deep learning structure for flame color specialization, $N{\times}N$ cell separation in detected flame area, flame image classification using deep learning structure for flame shape specialization. First, only the color of the flame is extracted from the input image and then labeled to detect the flame area. Second, area of flame detected is the input of a deep learning structure specialized in flame color and is classified as flame image only if the probability of flame class at the output is greater than 75%. Third, divide the detected flame region of the images classified as flame images less than 75% in the preceding section into $N{\times}N$ units. Fourthly, small cells divided into $N{\times}N$ units are inserted into the input of a deep learning structure specialized to the shape of the flame and each cell is judged to be flame proof and classified as flame images if more than 50% of cells are classified as flame images. To verify the effectiveness of the proposed deep learning structure, we experimented with a flame database of ImageNet. Experimental results show that the proposed deep learning structure has an average resource occupancy rate of 29.86% and an 8 second fast flame detection time. The flame detection rate averaged 0.95% lower compared to the existing deep learning structure, but this was the result of light construction of the deep learning structure for application to embedded systems. Therefore, the deep learning structure for flame detection proposed in this paper has been proved suitable for the application of embedded system.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1473-1475
• /
• 1996

In this paper, we propose the method to find the active region of flame which is produced within the gap between an oxygen blown converter and a skirt. For real-time image processing, basic region segmentation algorithms such as thresholding and XORing are used to segment the active region of flame. The result of this processing may be useful to clear the relationship between hood pressure and flame.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.23 no.1
• /
• pp.114-122
• /
• 1999

Reignition as special cases of acoustic pressure responses of flame are numerically studied by employing methanol droplet flame as a laminar flamelet. Quasi-steady flame responses occur in the range of small amplitude, low frequency oscillation. Reignition phenomena can occur when, by increasing the frequency of large amplitude acoustic pressure, the magnitude of characteristic acoustic time is the same order of that of characteristic reaction time of flames. And more increasing of amplitude of acoustic pressure induces the direct extinction of flame. Flame can sustain its own intensity even under the steady extinction temperature in case of high frequency acoustic oscillation, and this tendency is remarkable with increasing frequency. Reignition regime with respect to amplitude and frequency of acoustic pressure doesn't exist in low frequency($10^2$ Hz, in this study), but broadens with frequency of acoustic pressure.

• Journal of Mechanical Science and Technology
• /
• v.17 no.1
• /
• pp.139-145
• /
• 2003

Diode laser sensor is conducted to measure the gas temperature in the liquid-gas 2-phase counter flow flame. C$\_$10/H/ sub 22/ and city gas were used as liquid fuel and gas fuel, respectively. Two vibrational overtones of H$_2$O were selected and measurements were carried out in the spray flame region stabilized the above gaseous premixed flame. The path-averaged temperature measurement using diode laser absorption method succeeded in the liquid fuel combustion environment regardless of droplets of wide range diameter. The path-averaged temperature measured in the post flame of liquid-gas 2-phase counter flow flame showed qualitative reliable results. The successful demonstration of time series temperature measurement in the liquid-gas 2-phase counter flow flame gave us motivation of trying to establish the effective control system in practical combustion system. These results demonstrated the ability of real-time feedback from combustor inside using the non-intrusive measurement as well as the possibility of application to practical combustion system. Failure case due to influence of spray flame was also discussed.