• KIEAE Journal
• /
• v.7 no.5
• /
• pp.29-34
• /
• 2007

Discomfort glare from windows is an important issue in window and daylighting design. This study aims to investigate the effect of luminance difference between the lower and the upper part of a large glare source. Experiments were conducted using a luminous body divided into two parts, the upper part and the lower part. The degree of discomfort glare from the luminous body was examined. The result shows that the part with lower luminance is perceived as a glare source when the part has higher luminance then the background luminance. The degree of discomfort glare was estimated for the experimental conditions, and compared to the observations. The estimation was made using two methods: one summed the glare sensation for each section, and the other used the average luminance of the whole luminous body. The result of the comparison shows that the method using the average luminance has approximate values to the observations. Consequently, the use of the average luminance was proposed for evaluation of discomfort glare from non-uniform large glare sources.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.4
• /
• pp.394-400
• /
• 2000

The main goal of synchronization is to guarantee exclusive access to shared data and critical sections, and then it makes parallel programs work correctly and reliably. Exclusive access restricts parallelism of parallel programs, therefor efficient synchronization is essential to achieve high performance in shared-memory parallel programs. Many techniques are devised for efficient synchronization, which utilize features of systems and applications. This paper shows the simulation results that existing synchronization methods have inefficiency under CC-NUMA(Cache Coherent Non-Uniform Memory Access) system, and then compares the performance of Freeze&Melt synchronization that can remove the inefficiency. The simulation results present that Test-and-Test&Set synchronization has inefficiency caused by broadcast operation and the pre-defined order of Queue-On-Lock-Bit (QOLB) synchronization to execute a critical section causes inefficiency. Freeze&Melt synchronization, which removes these inefficiencies, has performance gain by decreasing the waiting time to execute a critical section and the execution time of a critical section, and by reducing the traffic between clusters.

• Journal of the Society of Naval Architects of Korea
• /
• v.53 no.4
• /
• pp.300-306
• /
• 2016

The gas ventilated by supercavitation splits into smaller bubbles and follows the water passage of the cavitation tunnel. The bubbles quickly return to the test section by rather high speed flow, and interrupt the observation of the supercavitation. To secure clear observation in the test section, the bubble collecting section(settling chamber) of large volume is prepared to collect bubbles in the water passage ahead of the test section. The bubble collecting section should provide enough buoyancy effect to the bubbles for proper bubble collecting. However, rather high-speed oncoming flow produces non-uniform velocity distribution and deteriorates buoyancy effect in the bubble collecting section. In the present study, the bubble collecting space and three porous plates are designed and analyzed through numerical methods, and the bubble collecting function is experimentally validated by 1/10-scaled model in terms of the formation of uniformly low velocity distribution in the bubble collecting section.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.12
• /
• pp.8654-8664
• /
• 2015

Beam-type structures with non-uniform cross sections are widely used in mechanical, architectural, and civil engineering fields. This paper deals with dynamic characteristics and vibration problems. Governing equations are first derived by using local coordinates. Their solutions are then assumed by using Galerkin's mode summation method. Bisection method is also applied in solving the determinant of the matrix which can provide natural frequencies. Whereas finite element methods adopt admissible functions satisfying only geometric boundary condition, in this study we apply Galerkin's mode summation method which uses eigen-functions satisfying both governing equations and boundary conditions. Modal analysis and experimental tests are finally performed using simply-supported beams with four different non-uniform cross-sections. Our analytical results then show good agreement with experimental ones.

• Steel and Composite Structures
• /
• v.16 no.1
• /
• pp.45-58
• /
• 2014

EC3 provides several methodologies for the stability verification of members and frames. However, when dealing with the verification of non-uniform members in general, with tapered cross-section, irregular distribution of restraints, non-linear axis, castellated, etc., several difficulties are noted. Because there are yet no guidelines to overcome any of these issues, safety verification is conservative. In recent research from the authors of this paper, an Ayrton-Perry based procedure was proposed for the flexural buckling verification of web-tapered columns. However, in order to apply this procedure, Linear Buckling Analysis (LBA) of the tapered column must be performed for determination of the critical load. Because tapered members should lead to efficient structural solutions, it is therefore of major importance to provide simple and accurate formula for determination of the critical axial force of tapered columns. In this paper, firstly, the fourth order differential equation for non-uniform columns is derived. For the particular case of simply supported web-tapered columns subject to in-plane buckling, the Rayleigh-Ritz method is applied. Finally, and followed by a numerical parametric study, a formula for determination of the critical axial force of simply supported linearly web-tapered columns buckling in plane is proposed leading to differences up to 8% relatively to the LBA model.

• Smart Structures and Systems
• /
• v.19 no.4
• /
• pp.431-439
• /
• 2017

This investigation is aimed to develop a model of experimental-computation determination of a support moment of a cantilever beam loaded with concentrated force at its end including the optimal choice of coordinates of deflection data points and parameters of transformation of deflection data in case of insufficient accuracy of the assignment of initial parameters (support settlement, angle of rotation of the bearing section) and cantilever beam length. The influence of distribution and characteristics of sensors on the cantilever beam on the accuracy of determining the support moment which improves in the course of transition from the uniform distribution of sensors to optimal non-uniform distribution is shown. On the basis of the theory of inverse problems the method of transformation reduction at numerical differentiation of deflection functions has been studied. For engineering evaluation formulae of uncertainty estimate to determine a support moment of a cantilever beam at predetermined uncertainty of measurements using sensors have been obtained.

• The KSFM Journal of Fluid Machinery
• /
• v.13 no.4
• /
• pp.45-50
• /
• 2010

The goal of paper is to investigate the flow and scalar distribution through the HP Nozzle Guide Vane (NGV) passage. Flow and scalar distribution measurement are conducted by using 5-hole pressure probe and $CO_2$ tracing technique, respectively. Three different experimental cases are considered depending on cooling flow condition. The result shows that the vortical secondary flow patterns are observed clearly and these flow characteristics maintain through the NGV passage regardless of cooling flow injection. Compared to center region, the high axial velocity flow is observed near wall region due to cooling flow injection. Without cooling flow, the $CO_2$ (scalar) distribution becomes to be uniform quickly due to the strong flow mixing phenomenon. However, in cases of cooling flow, scalar distribution is significantly non-uniform.

• Journal of the Korean Society for Precision Engineering
• /
• v.17 no.1
• /
• pp.209-215
• /
• 2000

Precision forging technology that can control flow velocity of workpiece have been developed to minimize the amounts of machining. To get the uniform rib length, flow velocity distribution is needed to be estimated and controlled. Computer-aided design is known for very effective to estimate the deformation behavior and design the die for controlling the flow velocity. In this study, die design to control the deformation velocity are investigated using the DEFORM-2D about rib-web shape parts. Also we can get uniform rib length by enforcing the back pressure at end section of rib. The applied load of back pressure farming is lower than that of conventional forging. These results are analysed and confirmed by the experiment.

• Earthquakes and Structures
• /
• v.23 no.2
• /
• pp.183-196
• /
• 2022

Reduced beam section (RBS) moment connections used in special moment resisting frames are currently limited to beam sections that are not larger than nominal depths of 920 mm, weight of 447 kg/m and flange thickness of 44 mm. Due to the higher demand for structural components with jumbo sections, which can potentially be applied in the transfer girders in long-span building structures, the newly available steel heavy members are promising. To address this issue, advanced numerical models are developed to fully evaluate the distribution of stresses and concentrations of plastic strains for such jumbo RBS connections. This paper first presents a brief overview of an experimental study on four specimens with large beam and column sections. Then, a numerical model that includes initial imperfections, residual stresses, geometric nonlinearity, and explicitly modeled welds is presented. The model is used to further explore the behavior of the test specimens, including distribution of stresses, distribution of plastic strains, stress triaxiality and potential for fracture. The results reveal that the stresses are highly non-uniform across the beam flange and, similarly, the plastic strains concentrate at the extreme fiber of the bottom flange. However, neither of these phenomena, which are primarily a function of beam flange thickness, is reflected in current design procedures.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.21 no.5
• /
• pp.615-627
• /
• 1997

An experiment was performed to local heat transfer coefficient and Nusselt number in the circular duct of 180.deg. bend for Re=6*10$^{4}$, 8*10$^{4}$ and 1*10$^{5}$ at swirling flow and non-swirling flow conditions. The test tube with circular section was made by stainless which has curvature ratio 9.4. The wall of test tube was heated directly by electrical power to 3.51 kw and swirling motion of air was produced by a tangential inlet to the pipe axis at the 180 degree. Measurements of local wall temperatures and bulk mean temperature of air are made at four circumferential positions in the 16 stations. The wall temperatures show particularly reduced distribution curve at bend for non-swirling flow but this effect does not appear for swirling flow. Nusselt number distributions for swirling flow which was calculated from the measured wall and bulk temperatures were higher than that of non-swirling flow. Average Nusselt number of swirling flow increased about 90 ~ 100% than that of non-swirling flow whole through the test tube. The Nu/N $u_{DB}$ values at the station of 90.deg. for non-swirling flow and swirling flow are respectively about 2.5 and 4.8 at Re=6*10$^{4}$. Also that is good agreement with Said's result for non-swirling flow. flow.