• Journal of Industrial Technology
• /
• v.28 no.B
• /
• pp.135-141
• /
• 2008

In this paper, the Proper Orthogonal Decomposition (POD) method, which is a statistical analysis technique to find the modal characteristics of a structure, is adapted to identify the modal parameters of a tall chimney structure. A wind force time history, which is applied to the structure, is obtained by a wind tunnel test of a scale down model. The POD method is applied on the wind force induced responses of the structure, and the true normal modes of the structure can be obtained. The modal parameters including, natural frequency, mode shape, damping ratio and kinetic energy of the structure can be estimated accurately. With these results, it may be concluded that the POD method can be applied to obtain accurate modal parameters from the wind-induced building responses.

• 한국가시화정보학회:학술대회논문집
• /
• 2005.12a
• /
• pp.46-49
• /
• 2005

Vents at outer walls of a large factory building are very important for natural ventilation. But, if a full-open vent is used, rain comes through the vents. We tried to utilize the natural ventilation effectively using a louver. A 1/120 scale-down building model was placed inside an atmospheric boundary layer simulated in a wind tunnel test section. The effect of louver angle on the ventilation flow inside the factory building was investigated experimentally. Instantaneous velocity fields inside the building model were measured using a 2-frame PIV system with varying the louver angles ($\theta=20^{\circ},\;40^{\circ},\;60^{\circ}$). For the case of $\theta=60^{\circ}$, as the incoming flow into the factory building increases, the inside velocity distribution becomes uniformly.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.17 no.11
• /
• pp.1050-1059
• /
• 2005

Heat and $SF_6$ gas dispersions over a complex terrain were investigated using wind tunnel. The wind speed, temperature and concentration profiles were measured for the 1/1000 scale complicated terrain model in an Eiffel type boundary layer wind tunnel with test section of 2.5m in height and 4.5m in width. The scale model was mounted on the top of a plate which can rotate with respect to the approaching wind. Dispersion processes from a continuous emission source driven by various wind direction were investigated, including plume climbing over the steep up-slope of the mountain and down-spreading toward the lower level of the valley. Extensive dispersion experiment data (wind speeds and concentration profiles) were provided for verification and validation of dispersion models. Under the identical flow and emission conditions, the independently measured profiles of the temperature and $SF_6$ concentration showed an excellent agreement which ensured the credibility of the results.

• Journal of the Korean GEO-environmental Society
• /
• v.13 no.2
• /
• pp.103-112
• /
• 2012

The most of original horizontal bearing capacity theory of the pile is not efficiently to consider interaction between soil and pile because it is only to consider the earth pressure theory and separately the ground form pile. In recent, in order to improve the pile technology, it is necessary to confirm the real behaviour characteristics of pile under lateral load. Hence, to evaluate the behaviour characteristics of the single and group pile under lateral loads using the strain wedge model that could consider the interaction between soil and piles. Primarily, laboratory scale down model tests was carried out to predict the behaviour characteristics on real size piles using the strain wedge model. The comparative analyses between model test and numerical analysis for the evaluation of whole behaviour were conducted.

• Wind and Structures
• /
• v.38 no.2
• /
• pp.129-146
• /
• 2024

Aiming at the problem of non-stationary wind field simulation of downbursts, a non-stationary down-burst generation system was designed by adding a nozzle and program control valve to the inlet of the original wall jet model. The computational fluid dynamics (CFD) method was used to simulate the downburst. Firstly, the two-dimensional (2D) model was used to study the outflow situation, and the database of working conditions was formed. Then the combined superposition of working conditions was carried out to simulate the full-scale measured downburst. The three-dimensional (3D) large eddy simulation (LES) was used for further verification based on this superposition condition. Finally, the wind tunnel test is used to further verify. The results show that after the valve is opened, the wind ve-locity at low altitude increases rapidly, then stays stable, and the wind velocity at each point fluctuates. The velocity of the 2D model matches the wind velocity trend of the measured downburst well. The 3D model matches the measured downburst flow in terms of wind velocity and pulsation characteris-tics. The time-varying mean wind velocity of the wind tunnel test is in better agreement with the meas-ured time-varying mean wind velocity of the downburst. The power spectrum of fluctuating wind ve-locity at different vertical heights for the test condition also agrees well with the von Karman spectrum, and conforms to the "-5/3" law. The vertical profile of the maximum time-varying average wind veloci-ty obtained from the test shows the basic characteristics of the typical wind profile of the downburst. The effectiveness of the downburst generation system is verified.

• Tunnel and Underground Space
• /
• v.14 no.1
• /
• pp.54-68
• /
• 2004

In this study, scaled model tests were performed on blasting demolition of reinforced concrete structures and the experimental results were analyzed in comparison with the results of numerical analysis. The tests were designed to induce a progressive collapse, and physical properties of the scaled model were determined using scale factors obtained ken dimension analysis. The scaled model structure was made of a mixture of plaster, sand and water at the ratio determined to yield the best scaled-down strength. Lead wire was used as a substitute for reinforcing bars. The scaled length was at the ratio of 1/10. Selecting the material and scaled factors was aimed at obtaining appropriately scaled-down strength. PFC2D (Particle Flow Code 2-Dimension) employing DEM (Distinct Element Method) was used for the numerical analysis. Blasting demolition of scaled 3-D plain concrete laymen structure was filmed and compared to results of numerical simulation. Despite the limits of 2-D simulation the resulting demolition behaviors were similar to each other. Based on the above experimental results in combination with bending test results of RC beam, numerical analysis was carried out to determine the blasting sequence and delay times. Scaled model test of RC structure resulted in remarkably similar collapse with the numerical results up to 900㎳ (mili-second).

• Journal of the Korean Society for Railway
• /
• v.16 no.1
• /
• pp.47-51
• /
• 2013

In this study, we carried out an experimental shield TBM excavation model test using a down-scale device in soft clay, to understand tunnel-face stability properties in relation to changes in slurry pressure. We performed five tests according to tunnel depth (0.5D, 0.75D, 1.0D, 1.25D, 1.5D), and compared theoretical tunnel-face pressure with model test results. The range in theoretical tunnel-face slurry pressure ($P_{min}{\leq}P_{slurry\;pressure}{\leq}P_{max}$), which is determined by earth pressure and water level, was very similar to the model test result. This result was due to the more isotropic condition of the soft clay ground, than of rocky ground.

• Geophysics and Geophysical Exploration
• /
• v.12 no.4
• /
• pp.355-360
• /
• 2009

This paper presents mobile robot based down-scale mineral resources exploration test system for the USN (Ubiquitous Sensor Network) based exploration. The system emulates the actual exploration environment. Underneath the metal free test plate, a metal object is attached. A magneto-meter mounted mobile robot runs around on the plate to find the metal. The measured magneto-meter values are transferred to the host PC via wireless network. The system enables to improve the reliability of simulation as well as to help efficient exploration system design. Metal-detecting experiments were carried out to illustrate the efficiency of the proposed system.

• Proceedings of the KSR Conference
• /
• 2007.05a
• /
• pp.430-437
• /
• 2007

The experiments for aerodynamic characteristics on railway acoustic screen are performed using 1/61 scale-down mock-up facility. The train model for the experiment is high speed train (KTX) and the tested speed is about 300km/h. The tested train length is about 61cm which is corresponding to two units of train. The strength of pressure wave is measured using piezo typed pressure sensor. The measured pressure is compared with field test data and UIC 779-1 values.

• Nutrition Research and Practice
• /
• v.13 no.6
• /
• pp.521-528
• /
• 2019

BACKGROUND/OBJECTIVES: The aim of this study was to develop Korean food image detection and recognition model for use in mobile devices for accurate estimation of dietary intake. MATERIALS/METHODS: We collected food images by taking pictures or by searching web images and built an image dataset for use in training a complex recognition model for Korean food. Augmentation techniques were performed in order to increase the dataset size. The dataset for training contained more than 92,000 images categorized into 23 groups of Korean food. All images were down-sampled to a fixed resolution of $150{\times}150$ and then randomly divided into training and testing groups at a ratio of 3:1, resulting in 69,000 training images and 23,000 test images. We used a Deep Convolutional Neural Network (DCNN) for the complex recognition model and compared the results with those of other networks: AlexNet, GoogLeNet, Very Deep Convolutional Neural Network, VGG and ResNet, for large-scale image recognition. RESULTS: Our complex food recognition model, K-foodNet, had higher test accuracy (91.3%) and faster recognition time (0.4 ms) than those of the other networks. CONCLUSION: The results showed that K-foodNet achieved better performance in detecting and recognizing Korean food compared to other state-of-the-art models.