• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.11
• /
• pp.2432-2438
• /
• 1997

This paper proposes recognition system using multi-backpropagation neural networks rather than single backpropagation neural network to enhance the rate of character recognition resultsing from extracting the region of velhicle number in that the image of vehicle number plate from CCD camera has a distinguish feature, that is, illumination of a pattern. The experiment in this paper shows an output that the method using multi-backpropagation neural networks rather than signal backpropagation neural network takes less training time for computation and also has higher recognition rage of vehicle number.

• Proceedings of the KSME Conference
• /
• 2002.08a
• /
• pp.237-240
• /
• 2002

In the present study, we perform LES of turbulent flow and temperature fields in a circular impinging jet at Re=5000 for two cases of H/D=2 and 6 (H denotes the distance between the jet exit and flat plate, and D does the diameter of the jet exit). In the case of H/D=2, the regular vortical structures observed in free jet do not exist because of the smaller distance than the potential core. The Nusselt number on the wall is largest at $r/D{\cong}10.67$ where vortex rings Impinge. At $r/D{\cong}1.5{\~}2.0$, the vortex rings induce the secondary vortices, resulting in a secondary peak in the Nusselt number there. In the case of H/D=6, the vortex rings change into three-dimensional vortical structures and the small-scale vortices impinge on the flat plate. The increase of turbulent intensity due to small-scale vortices results in the largest Nusselt number at the stagnation point.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.20 no.4
• /
• pp.309-332
• /
• 2016

Finite element method has been applied to solve the fundamental governing equations of natural convective, electrically conducting, incompressible fluid flow past an infinite vertical plate surrounded by porous medium in presence of thermal radiation, viscous dissipation, Soret and Dufour effects. In this research work, the results of coupled partial differential equations are found numerically by applying finite element technique. The sway of significant parameters such as Soret number, Dufour number, Grashof number for heat and mass transfer, Magnetic field parameter, Thermal radiation parameter, Permeability parameter on velocity, temperature and concentration evaluations in the boundary layer region are examined in detail and the results are shown in graphically. Furthermore, the effect of these parameters on local skin friction coefficient, local Nusselt number and Sherwood numbers is also investigated. A very good agreement is noticed between the present results and previous published works in some limiting cases.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.7
• /
• pp.1675-1683
• /
• 1995

A numerical study of the fluid flow and heat transfer characteristics of the two-dimensional impingement jet with a confinement plate has been carried out. The fluid flow was calculated by solving the full Navier-Stokes equation. In doing that, the well known SIMPLER algorithm was used and the trouble making convection term was discretized according to QUICKER scheme. The energy equation was simply solved by using the SOR method. For the Reynolds number of 10000, two channel heights, say 1.5 and 3.0 times the jet exit width, and two thermal boundary conditions constant wall temperature and constant wall heat flux were considered. Discrete heat sources were flush mounted along the impingement plate at a distance of 0, 2, 3, 4, 5, 6, 10, 12, times the jet exit width from the stagnation point. The length of each heat source is 4 times the jet exit width long. The Nusselt number averaged over each heat source was compared with experiment. Comparison shows that both calculations and experiment have the secondary peak of Nusselt number at downstream of stagnation point, even though there is a little quantitative difference in between. The difference is believed due to abscure thermal boundary condition in experiment and also accuracy of turbulence model used. The secondary peak is shown to be caused by rigorous turbulent flow motion generated as the wall jet flow is retarded and developes into the channel flow without flow reversal.

• Journal of Energy Engineering
• /
• v.10 no.3
• /
• pp.272-277
• /
• 2001

An experimental study of jet impinging the non-isothermal heating surface with linear temperature gradient is conducted with the presentation of the turbulent flow characteristics and the heat transfer rate, represented by the Nusselt number. The jet Reynolds number ranges from 15,000 to 30,000, the temperature gradient of the plate is 2~4.2$^{\circ}C$/cm and the dimensionless nozzle to plate distance (H/D) is from 2 to 10. The results show that the peak of heat transfer rate occurs at the stagnation point, and the heat transfer rate decreases as the radial distance from the stagnation point increases. A remarkable feature of the heat transfer rate is the existence of the second peak. This is due to the turbulent development of the wall jet. Maximum heat transfer rate occurs when the axial distance from the nozzle to nozzle diameter (H/D) is 6 or 8. The heat transfer rate can be correlated as a power function of Prandtl number, Reynolds number, the dimensionless nozzle to plate distance (H/D) and temperature gradient (dT/dr). It has been found that the heat transfer rate increases with increasing turbulent intensity. The wall jet is influenced by temperature gradient and the effect becomes more important at higher radii.

• KCI Concrete Journal
• /
• v.14 no.1
• /
• pp.51-60
• /
• 2002

Strengthening of existing concrete structures is a major concern in recent years as the number of degraded structures increases. The purpose of this paper is to investigate the static and fatigue behavior of reinforced concrete (RC) beams strengthened with steel plates. To this end, a comprehensive test program has been set up and many series of strengthened beams have been tested. The major test variables include the plate thickness, adhesive thickness, and the shear-span to depth ratio. The test results indicate that the separation of plates is the dominant failure mechanism even for the full-span-length strengthened beams with steel plate. The theoretical ultimate load capacities for strengthened beams based on the full composite action of concrete beam and steel plate are found to be larger than the actual measured load capacities. The strengthened beams exhibit more dominant shear cracking as the shear-span to depth ratio decreases. The ultimate capacity of strengthened beams increases slightly with the increase of adhesive thickness, which may be caused by the late initiation of plate separation in the beams with thicker adhesive. A realistic concept of ductility for plate-strengthened beams is proposed in this study. It is seen that the strengthened beams show relatively low ductility compared with unstrengthened beams. The present study indicates that the strengthened beams exhibit much higher fatigue resistance than the unstrengthened beams. The increase of deflections of strengthened beams according to the number of load cycles is much smaller than that of unstrengthened beams. The present study provides very useful results for the realistic application of plate-strengthening method in reinforced concrete structures.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.4 no.4
• /
• pp.243-252
• /
• 1992

A mixed convection heat transfer from two vertical parallel plates has been studied numerically by the finite difference method. Effects of the Grashof number, the relative length, $L_2/L_1$. the dimensionless temperature ratio, ${\Phi}_2/{\Phi}_1$ and the dimensionless plate spacing, $b/L_1$ are examined for the heat transfer. Independent of the Grashof numbers and $L_2/L_1$, the dimensionless vertical velocity distributions skewed on the left plate as ${\Phi}_2/{\Phi}_1$ decreased. The dimensionless vertical velocity distribution for $Gr/Re^2=1$ and ${\Phi}_2/{\Phi}_1=1.0$ is skewed to the right plate $L_2/L_1=0.5$, symmetric at $L_2/L_1=1.0$ and skewed to the left plate at $L_2/L_1=1.5$. But for $Gr/Re_2=10.0$ and ${\Phi}_2/{\Phi}_1=1.0$ reversed velocity patterns are obtained. Regardless of the Grashof numbers and $L_2/L_1$, the mean Nusselt nembers on the inside surface of the left plate decreases and those of the right inside surface increases as ${\Phi}_2/{\Phi}_1$ increases. Temperature, velocity and mean Nusselt number distributions are apparently not affected by $L_2/L_1$.

• Steel and Composite Structures
• /
• v.37 no.1
• /
• pp.91-98
• /
• 2020

Steel plate shear walls are recently used as efficient seismic lateral resisting systems. These lateral resistant structures are implemented to provide more strength, stiffness and ductility in limited space areas. In this study, the seismic behavior of the multi-story steel frames with steel plate shear walls are investigated for buildings with 4, 8, 12 and 16 stories using verified computational modeling platforms. Different number of steel moment bays with distinctive lengths are investigated to effectively determine the deflection amplification factor for low-rise and high-rise structures. Results showed that the dissipated energy in moment frames with steel plates are significantly related to the inside panel. It is shown that more than 50% of the dissipated energy under various ground motions is dissipated by the panel itself, and increasing the steel plate length leads to higher energy dissipation capability. The deflection amplification factor is studied in details for various verified parametric cases, and it is concluded that for a typical multi-story moment frame with steel plate shear walls, the amplification factor is 4.93 which is less than the recommended conservative values in the design codes. It is shown that the deflection amplification factor decreases if the height of the building increases, for which the frames with more than six stories would have less recommended deflection amplification factor. In addition, increasing the number of bays or decreasing the steel plate shear wall length leads to a reduction of the deflection amplification factor.

• Geomechanics and Engineering
• /
• v.9 no.3
• /
• pp.373-395
• /
• 2015

Comprehensive results from cyclic plate loading at a diameter of 300 mm supported by layers of geocell are presented. The plate load tests were performed in a test pit measuring $2000{\times}2000mm$ in plane and 700 mm in depth. To simulate half and full traffic loadings, fifteen loading and unloading cycles were applied to the loading plate with amplitudes of 400 and 800 kPa. The optimum embedded depth of the first layer of geocell beneath the loading plate and the optimum vertical spacing of geocell layers, based on plate settlement, are both approximately 0.2 times loading plate diameter. The results show that installation of the geocell layers in the foundation bed, increase the resilient behavior in addition to reduction of accumulated plastic and total settlement of pavement system. Efficiency of geocell reinforcement was decreased by increasing the number of the geocell layers for all applied stress levels and number of cycles of applied loading. The results of the testing reveal the ability of the multiple layers of geocell reinforcement to 'shakedown' to a fully resilient behavior after a period of plastic settlement except when there is little or no reinforcement and the applied cyclic pressure are large. When shakedown response is observed, then both the accumulated plastic settlement prior to a steady-state response being obtained and the resilient settlements thereafter are reduced. The use of four layers of geocell respectively decreases the total and residual plastic settlements about 53% and 63% and increases the resilient settlement 145% compared with the unreinforced case. The inclusion of the geocell layers also reduces the vertical stress transferred down through the pavement by distributing the load over a wider area. For example, at the end of the load cycle of the applied pressure of 800 kPa, the transferred pressure at the depth of 510 mm is reduced about 21.4%, 43.9%, 56.1% for the reinforced bases with one, two, and three layers of geocell, respectively, compared to the stress in the unreinforced bed.

• International Journal of Air-Conditioning and Refrigeration
• /
• v.12 no.3
• /
• pp.158-167
• /
• 2004

Condensation heat transfer experiments were conducted with a oblong shell and plate heat exchanger without oil in a refrigerant loop using R-134a. An experimental refrigerant loop has been developed to measure the condensation heat transfer coefficient $h_r$ and frictional pressure drop ${\Delta}p_f$ of R-134a in a vertical oblong shell and plate heat exchanger. Four vertical counter flow channels were formed in the oblong shell and plate heat exchanger by four plates having a corrugated sinusoid shape of a $45^{\circ}$ chevron angle. The effects of the refrigerant mass flux, average heat flux, refrigerant saturation temperature and vapor quality were explored in detail. Similar to the case of a plate heat exchanger, even at a very low Reynolds number, the flow in the oblong shell and plate heat exchanger remains turbulent. The results indicate that the condensation heat transfer coefficients and pressure drops increase with the vapor quality. A rise in the refrigerant mass flux causes an increase in the $h_r\;and\;{\Delta}p_f$. Also, a rise in the average heat flux causes an increase in the $h_r$. But the effect of the average heat flux does not show significant effect on the ${\Delta}p_f$. On the other hand, at a higher saturation temperature, both the $h_r\;and\;{\Delta}p_f$. found to be lower. Based on the present data, the empirical correlations are provided in terms of the Nusselt number and friction factor.