• Journal of the Korean Institute of Gas
• /
• v.23 no.3
• /
• pp.20-26
• /
• 2019

Selective catalytic reduction(SCR) method is widely used among various methods for reducing nitrogen oxides in combustion devices of coal power plant. In the present study, the computational fluid dynamic analysis was accomplished to derive the optimal shape of ammonia-dilution air mixing device in a ammonia injection grid. The distribution characteristics of flow and $NH_3$ concentration had been elucidated for the reference shape of ammonia mixing device(Case 1). In the mixing device of Case 1, it could be seen that $NH_3$ distribution was shifted to the wall opposite to the inlet of the ammonia injection pipe. For the improvement of $NH_3$ distribution, the case(Case 2) with closing one upper injection hole and 4 side injection holes, the case(Case 3) with installing horizontal plate at the upper of ammonia injection pipe, the case(Case 4) with installing horizontal plate and horizontal arc plate at he upper of ammonia injection pipe were investigated by analyzing flow and $NH_3$ concentration distributions. From the present study, it was found that the % RMS of $NH_3$ for Case 4 was 4.92%, which was the smallest value among four cases, and the range of $R_{NH3}$ also has the optimally uniform distribution, -10.82~8.34%.

• Earthquakes and Structures
• /
• v.16 no.2
• /
• pp.221-234
• /
• 2019

Today, many important concrete face rockfill dams (CFRDs) have been built on the world, and some of these important structures are located on the strong seismic regions. In this reason, examination and monitoring of these water construction's seismic behaviour is very important for the safety and future of these dams. In this study, the nonlinear seismic behaviour of Ilısu CFR dam which was built in Turkey in 2017, is investigated for various reservoir water heights taking into account 1995 Kobe near-fault and far-fault ground motions. Three dimensional (3D) finite difference model of the dam is created using the FLAC3D software that is based on the finite difference method. The most suitable mesh range for the 3D model is chosen to achieve the realistic numerical results. Mohr-Coulomb nonlinear material model is used for the rockfill materials and foundation in the seismic analyses. Moreover, Drucker-Prager nonlinear material model is considered for the concrete slab to represent the nonlinearity of the concrete. The dam body, foundation and concrete slab constantly interact during the lifetime of the CFRDs. Therefore, the special interface elements are defined between the dam body-concrete slab and dam body-foundation due to represent the interaction condition in the 3D model. Free field boundary condition that was used rarely for the nonlinear seismic analyses, is considered for the lateral boundaries of the model. In addition, quiet artificial boundary condition that is special boundary condition for the rigid foundation in the earthquake analyses, is used for the bottom of the foundation. The hysteric damping coefficients are separately calculated for all of the materials. These special damping values is defined to the FLAC3D software using the special fish functions to capture the effects of the variation of the modulus and damping ratio with the dynamic shear-strain magnitude. Total 4 different reservoir water heights are taken into account in the seismic analyses. These water heights are empty reservoir, 50 m, 100 m and 130 m (full reservoir), respectively. In the nonlinear seismic analyses, near-fault and far-fault ground motions of 1995 Kobe earthquake are used. According to the numerical analyses, horizontal displacements, vertical displacements and principal stresses for 4 various reservoir water heights are evaluated in detail. Moreover, these results are compared for the near-fault and far-faults earthquakes. The nonlinear seismic analysis results indicate that as the reservoir height increases, the nonlinear seismic behaviour of the dam clearly changes. Each water height has different seismic effects on the earthquake behaviour of Ilısu CFR dam. In addition, it is obviously seen that near-fault earthquakes and far field earthquakes create different nonlinear seismic damages on the nonlinear earthquake behaviour of the dam.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.11
• /
• pp.435-442
• /
• 2018

Outlier detection means to detect data samples that deviate significantly from the distribution of normal data. Most outlier detection methods calculate an outlier score that indicates the extent to which a data sample is out of normal state and determine it to be an outlier when its outlier score is above a given threshold. However, since the range of an outlier score is different for each data and the outliers exist at a smaller ratio than the normal data, it is very difficult to determine the threshold value for an outlier score. Further, in an actual situation, it is not easy to acquire data including a sufficient amount of outliers available for learning. In this paper, we propose a clustering-based outlier detection method by constructing a model representing a normal data region using only normal data and performing binary classification of outliers and normal data for new data samples. Then, by dividing the given normal data into chunks, and constructing a clustering model for each chunk, we expand it to the ensemble method combining the decision by the models and apply it to the streaming data with dynamic changes. Experimental results using real data and artificial data show high performance of the proposed method.

• Journal of Broadcast Engineering
• /
• v.24 no.3
• /
• pp.472-484
• /
• 2019

This paper proposes a time-domain sound event detection algorithm using DNN (Deep Neural Network). In this system, time domain sound waveform data which is not converted into the frequency domain is used as input to the DNN. The overall structure uses CRNN structure, and GLU, ResNet, and Squeeze-and-excitation blocks are applied. And proposed structure uses structure that considers features extracted from several layers together. In addition, under the assumption that it is practically difficult to obtain training data with strong labels, this study conducted training using a small number of weakly labeled training data and a large number of unlabeled training data. To efficiently use a small number of training data, the training data applied data augmentation methods such as time stretching, pitch change, DRC (dynamic range compression), and block mixing. Unlabeled data was supplemented with insufficient training data by attaching a pseudo-label. In the case of using the neural network and the data augmentation method proposed in this paper, the sound event detection performance is improved by about 6 %(based on the f-score), compared with the case where the neural network of the CRNN structure is used by training in the conventional method.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.5
• /
• pp.627-635
• /
• 2019

A programmable drone is a drone developed not only to experience the basic principles of flight but also to control drones through Arduino-based programming. Due to the nature of the training drones, the main users are students who are inexperienced in controlling the drones, which often cause frequent collisions with external objects, resulting in high damage to the drones' frame. In this study, the structural stability of the drone was evaluated by means of a structural dynamics based collision simulation for educational drone frame. Collision simulations were performed on three cases according to the impact angle of $0^{\circ}$, $+15^{\circ}$ and $-15^{\circ}$, using an analytical model with approximately 240,000 tetrahedron elements. Using ANSYS LS-DYNA, which provides excellent functions for the simulation of the dynamic behavior of three-dimensional structures, the stress distribution and strain generated on the drone upper, the drone lower, and the ring assembly were analyzed when the drones collided against the wall at a rate of 4 m/s. Safety factors resulting from the equivalent stress and the yield strain were calculated in the range of 0.72 to 2.64 and 1.72 to 26.67, respectively. To ensure structural stability for areas where stress exceeds yield strain and ultimate strain according to material properties, the design reinforcement is presented.

• Korean Chemical Engineering Research
• /
• v.57 no.3
• /
• pp.358-367
• /
• 2019

Adsorption equilibrium and intraparticle diffusion characteristics of benzene, toluene, and xylene vapors on activated carbon and zeolite 13X were investigated. Static adsorption experiments were carried out under the pressure range of 0.01~0.07 bar while changing the adsorption temperature to 293.15 K, 303.15 K, and 313.15 K, respectively. Adsorption equilibrium was analyzed by Langmuir, Freundlich and Toth models. The adsorption energy was 5.26~31.0 kJ/mol representing physical adsorption characteristics. The maximum adsorption capacity on activated carbon was the largest for benzene, and the smallest for xylene. Toluene was in between. In the case of zeolite 13X, the maximum adsorption capacity was the largest for xylene, and the smallest for benzene as opposed to activated carbon. The effective diffusion coefficients of gas adsorbate were measured to be about $10^{-5}{\sim}10^{-4}cm^2/s$, and increased with temperature. As the pressure increased, the effective diffusion coefficients were decreased. The dependence of effective diffusion coefficients on temperature and pressure was greater in zeolite 13X particles than in activated carbon. Therefore, it is necessary to express the diffusion coefficients as a function of pressure in order to predict the precise dynamic behavior of the adsorption process using zeolite 13X where the pressure fluctuation occurs abruptly.

• Korean Journal of Applied Biomechanics
• /
• v.31 no.1
• /
• pp.1-15
• /
• 2021

Objective: The aim of this study was to investigate 1) the difference in static lower extremity alignment (SLEA) according to a history of lateral ankle sprain (LAS), 2) to identify SLEA factors affecting LAS, and 3) to present the cut-off value and 4) the usefulness and limitations of the SLEA measurement. Method: This case-control study recruited 88 men (age: 27.78±4.69 yrs) and 39 women (age: 24.62±4.20 yrs) subjects with and without LAS. SLEA measurement protocol included Q angle, tibiofemoral angle, genu recurvatum, rear foot (RF) angle, tibal varum and torsion, navicular drop, ankle dorsiflexion range of motion (DF ROM). Independent t-test, logistic regression and receiver operating characteristic (ROC) curve were used for statistical analysis. Results: Men with a history of LAS had significantly smaller Q angles both in standing and in supine position, while women with a history of LAS had significantly greater DF ROM in non-weight bearing (NWB; p < 0.05). Logistic regression model suggests tibial varum (OR = 0.779, p = 0.021) and WB DF ROM (OR = 1.067, p = 0.045) were associated with LAS in men. In case of women, there were no significant SLEA factors for LAS, however, ROC curve analysis revealed standing RF angle (AUC = 0.647, p = 0.028) and NWB DF ROM (AUC = 0.648, p = 0.026) could be affecting factors for LAS. Conclusion: There are differences in SLEA according to the history of LAS, furthermore, the identified items were different by sex. In case of men, tibial varum and WB DF ROM affect LAS occurrence. Standing RF angle and NWB DF ROM of women could be a predictor for LAS. However, since the sensitivity and specificity in most of the SLEA measurements are low, kinematic in dynamic tasks should be considered together for a more accurate evaluation of LAS risk.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_1
• /
• pp.1323-1337
• /
• 2020

GEO-KOMPSAT-2A AMI (Advanced Meteorological Imager) radiometric calibration evaluation is an essential element not only for functional and performance verification of the payload but for the quality of the sensor data. AMI instrument consists of six reflective channels and ten thermal infrared ones. One of the key parameters representing radiometric properties of the sensor is a SNR (Signal-to-Noise Ratio) for the reflective channels and a NEdT (Noise Equivalent delta Temperature) for the IR ones respectively. Other important radiometric calibration parameters are a dynamic range and a gain value related with the responsivity of detectors. To verify major radiometric calibration performance of AMI, an offline radiometric evaluation tool was developed separately with a real-time AMI data processing system. Using the evaluation tool, validation activities were carried out during the GEO-KOMPSAT-2A In-Orbit Test period. The results from the evaluation tool were cross checked with those of the HARRIS, which is the AMI payload vendor. AMI radiometric evaluation activities were conducted through three phases for both sides (Side 1 and Side 2) of AMI payload. Results showed that performances of the key radiometric properties were outstanding with respect to the radiometric requirements of the payload. The effectiveness of the evaluation tool was verified as well.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.28 no.4
• /
• pp.610-619
• /
• 2022

Hydrogen is emerging as an alternative fuel for eco-friendly ships because it reacts with oxygen to produce electrical energy and only water as a by-product. However, unlike regular fossil fuels, hydrogen has a material with a high risk of explosion due to its low ignition point and high flammability range. In order to safely use hydrogen in ships, it is an essential task to study the flow characteristics of hydrogen leakage and diffusion need to be studied. In this study, a numerical analysis was performed on the effect of leakage, ventilation, etc. on ventilation performance when hydrogen leaks in an enclosed space such as inside a ship. ANSYS CFX ver 18.1, a commercial CFD software, was used for numerical analysis. The leakage rate was changed to 1 q, 2 q, and 3 q at 1 q = 1 g/s, the ventilation rate was changed to 1 Q, 2 Q and 3 Q at 1 Q = 0.91 m/s, and the ventilation method was changed to type I, type II, type III to analyze the ventilation performance was analyzed. As the amount of leakage increased from 1 q to 3 q, the HMF in the storage room was about 2.4 to 3.0 times higher. Furthermore, the amount of ventilation to reduce the risk of explosion should be at least 2 Q, and it was established that type III was the most suitable method for the formation of negative pressure inside the hydrogen tank storage room.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.25 no.6
• /
• pp.50-59
• /
• 2021

Currently, the response correction factor is calculated by comparing the response measured by the load test on a bridge with the response analyzed in the initial analytical model. Then the load rating and the load carrying capacity are evaluated. However, the response correction factor gives a value that fluctuates depending on the measurement location and load condition. In particular, when the initial analytical model is not suitable for representing the behavior of a bridge, the range of variation is large and the analysis response by the calibrated model may give a result that is different from the measured response. In this study, a pseudo-static load test was applied to obtain static response with dynamic components removed under various load conditions of a vehicle moving at a low speed. Static response was measured on two similar PSC-I girder bridges, and the response correction factors for displacement and strain were calculated for each of the two bridges. When the initial analysis model was not properly set up, it is verified that the response of the analytical model corrected by the average response correction factor does not fall within the margin of error with the measured response.