• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.733-745
• /
• 2017

Local composite quantile regression is a useful non-parametric regression method widely used for its high efficiency. Data smoothing methods using kernel are typically used in the estimation process with performances that rely largely on the smoothing parameter rather than the kernel. However, $L_2$-norm is generally used as criterion to estimate the performance of the regression function. In addition, many studies have been conducted on the selection of smoothing parameters that minimize mean square error (MSE) or mean integrated square error (MISE). In this paper, we explored the optimality of selecting smoothing parameters that determine the performance of non-parametric regression models using local linear composite quantile regression. As evaluation criteria for the choice of smoothing parameter, we used mean absolute error (MAE) and mean integrated absolute error (MIAE), which have not been researched extensively due to mathematical difficulties. We proved the uniqueness of the optimal smoothing parameter based on MAE and MIAE. Furthermore, we compared the optimal smoothing parameter based on the proposed criteria (MAE and MIAE) with existing criteria (MSE and MISE). In this process, the properties of the proposed method were investigated through simulation studies in various situations.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.32 no.3A
• /
• pp.149-160
• /
• 2012

Corrugated steel plates have several advantages such as high resistance for shear without stiffeners, minimization of welding process, and high fatigue resistance. To take advantage of these benefits, several researchers have attempted to use corrugated steel plate as a web of I-girders. The lateral-torsional buckling is the major design aspect of such I-girders. However, lateral-torsional buckling of the I-girder with corrugated steel webs still needs to be investigated especially for a real loading condition such as non-uniform bending. This paper investigated the lateral-torsional buckling strength of the I-girder with corrugated steel webs under linear moment gradient by using finite element analysis. From the results, it was found that the buckling behavior of the I-girder with corrugated steel webs differed depending on the number of periods of the corrugation. Also, a simple equation for the moment gradient correction factor of the I-girder with corrugated steel webs was suggested. The inelastic lateral-torsional buckling strength of the I-girder with corrugated steel webs was then discussed based on current design equations for ordinary I-girders and the results of finite element analysis.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.10 no.3 s.49
• /
• pp.85-100
• /
• 2006

This paper proposes an optimal design method of linear viscous dampers for the seismic performance of two adjacent structures with different heights. Accordingly, connection method using diagonal bracing between two floors and connection method between two structures are considered, and the effectiveness of the latter method is confirmed through the comparison of the frequency response functions with respect to damping capacity. Moreover, optimal damping to minimize the response of the adjacent structures in the frequency domain is found. The sensitivity of natural frequency and modal damping according to the damper capacity at each floor is obtained for the optimally designed system. From the sensitivity analysis, the modal damping is evaluated to be very sensitive to the damper installed at higher floor. Therefore, sensitivity-based damping distribution method is proposed. Diagonal bracing connection method, uniform distribution method and sensitivity-based distribution method are compared to each other in terms of seismic performance. The comparative results demonstrate that the proposed method is an effective seismic design method for the adjacent structures.

• Proceedings of the Korean Institute of Surface Engineering Conference
• /
• 2003.05a
• /
• pp.55-55
• /
• 2003

최근 높은 해상도의 평판 디스플레이 장치 특히 차세대 TFT-LCD를 개발하기 위해서는 건식식각공정의 개발이 필수 불가결하며 이는 플라즈마 공정장치의 대면적화가 가능해야 한다. 따라서 산업계는 이러한 제조 조건에 알맞는 대면적 플라즈마 반응기 개발을 추구하고 있다. 이를 위해서는 건식식각공정의 개발이 필수 불가결하며 이를 위해선 플라즈마 공정장 치의 대면적화가 가능해야 한다. 이러한 대면적 공정을 위해서는 낮은 공정압력, 고밀도, 높은 플라즈마 균일도가 요구된다. 또한 이러한 대면적 고밀도 플라즈마에의 적용을 위하여 새로운 유도결합형 플라즈마 소오스의 개발이 진행되고 있으며, 안정적인 300mm웨이퍼 공정을 위하여 여러 형태의 안테나가 연구되어지고 있다. 그러나 차세대 TFT-LCD에 적용 가 능하게끔 기존의 ICP 소오스를 직접적으로 대면적화 하는데 있어서는 안테나의 인덕턴스의 값이 키지며, 유전물질의 두께 증가 및 그에 따른 재료비의 상슴에 의해 그 한계점을 나타 내었다. 본 연구에서는 차세대 TFT-LCD 및 POP 대면적 공정에 적용 가능한 고밀도 플라즈마를 발생시키기 위해서 내장형 유도결합형 선형 안테나를 사용하였다. 내장형 유도결합형 선형 안테나가 가지고 있는 고유의 정전기적 결합효과를 최소화시키기 위해 직사각형모양의 플라즈마 챔버(830mm*1,020mm)에서 영구자석을 사용하였다. 영구자석을 사용하여 외부자 장을 인가하였을 때가, 그럴지 않은 때보다 RF 안테나에 걸리는 코일의 전압을 낮춰주었으며, 영구자석의 배열에 따라 코일의 인덕턴스의 값이 크게 변함을 알 수 있었다. 그리고, 최적화된 자장의 배열은 플라즈마의 이온밀도를 증가시켰으며, 플라즈마 균일도 또한 10% 이 내로 유지됨을 알 수 있었다. 따른 식각 메커니즘에 대하여 알아보고자 하였다. $CF_4/Cl_2$ gas chemistry 에 첨 가 가스로 $N_2$와 Ar을 첨 가할 경 우 텅 스텐 박막과 하부 layer 간의 etch selectivity 증가는 관찰되지 않았으며, 반면에 첨가 가스로 $O_2$를 사용할 경우, $O_2$의 첨가량이 증가함에 따라 etch s selectivity 는 계속적으로 증가렴을 관찰할 수 있었다. 이는 $O_2$ 첨가에 따라 형성되는 WOF4 에 의한 텅스텐의 etch rates 의 감소에 비하여, $Si0_2$ 등의 형성에 의한 poly-Si etch rates 이 더욱 크게 감소하였기 때문으로 사료된다. W 과 poly-Si 의 식각 특성을 이해하기 위하여 X -ray photoelectron spectroscopy (XPS)를 사용하였으며, 식각 전후의 etch depth 를 측정하기 위하여 stylus p pmfilometeT 를 이용하였다.X> 피막이 열처리 전후에 보아는 기계적 특성의 변화 양상은 열역학적으로 안정한 Wurzite-AlN의 석출에 따른 것으로 AlN 석출상의 크기에 의존하며, 또한 이러한 영향은 $(Ti_{1-x}AI_{x})N$ 피막에 존재하는 AI의 함량이 높고, 초기에 증착된 막의 업자 크기가 작을 수록 클 것으로 여겨진다. 그리고 환경의 의미의 차이에 따라 경관의 미학적 평가가 달라진 것으로 나타났다.corner$적 의도에 의한 경관구성의 일면을 확인할수 있지만 엄밀히 생각하여 보면 이러한 예의 경우도 최락의 총체적인 외형은 마찬가지로

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.16 no.6 s.97
• /
• pp.603-614
• /
• 2005

In this paper, digital ultra-narrowband Walky-Talky using direct conversion method for CQPSK modulation scheme is implemented with satisfying the requirements of APCO P25. RF transceiver design and implementation scheme that minimize the influence of DC-offset and AC-coupling at ultra-narrowband is proposed. This scheme also minimizes the influence of nonlinear characteristic at power amplifier fir CQPSK modulation method. Test results of full system including DSP module and direct conversion RF transceiver show that FCC emission mask at 36.8 dBm PEP meets the standard requirements. The characteristic of receiver AGC by PWM control signal is linear at 40 dB dynamic range and voice communication at input power level of -116 dBm is successful. Also it is verified that the performance of BER versus frequency offset and versus SNR meets the standard requirements.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.1
• /
• pp.61-67
• /
• 2000

Digital controllers found in many industrial real-time systems consist of a number of interacting periodic tasks. To sustain the required control quality, these tasks possess the maximum activation periods as performance constraints. An essential step in developing a real-time system is thus to assign each of these tasks a constant period such that the maximum activation requirements are met while the system utilization is minimized [1]. Given a task graph design allowing producer/consumer relationships among tasks [2], resource demands of tasks, and range constraints on periods, the period assignment problem falls into a class of nonlinear optimization problems. This paper proposes a ploynomial time approximation algorithm which produces a solution whose utilization does not exceed twice the optimal utilization. Our experimental analysis shows that the proposed algorithm finds solutions which are very close to the optimal ones in most cases of practical interest.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.35 no.3
• /
• pp.309-316
• /
• 2011

Canoes are usually made from wood or FRP. However, today environment-friendly materials are preferred, and hulls made of FRP are prohibited in some countries. Polyethylene can be recycled and so is suitable for synthetic canoe construction. We used 3D Boat-Design to determine the hydrostatic properties of the canoe. Flow-structure coupled analysis was performed using ANSYS Workbench R12.1. The hull pressure and passenger weight were considered as canoe loading factors. The key parameters for the canoe are the design variables. The constraints are as follows: (1) The maximum stress must not exceed 50% of the polyethylene yield stress; and (2) the canoe weight must not exceed 50 kg. The optimal structural conditions were obtained by the response optimization process. The components of the canoe hull were manufactured from polyethylene pipes and joined by thermal fusion methods. Tests showed that the polyethylene canoe had better performance than existing canoes.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.35 no.3
• /
• pp.589-597
• /
• 2015

Wave energy harvesting system using OWC(Oscillating Water Column) and MB (Movable Body) attached at the caisson breakwater was studied. This system was suggested to maximize wave energy extraction using resonant phenomena of oscillating water column and buoy in wave channel (Park et al., 2014). Not only incident waves but also reflected waves from the breakwater can be used as sources of exciting force for harvesting wave energy efficiently. Using Galerkin finite model based on the linear wave theory (Park, 1991), the performance of the system was evaluated for various damping ratios of power take off system. Numerical results show that the proposed system is excellent in efficiency compared with that of conventional system and the performance of the system is governed by the resonance of oscillating water column in the wave channel. In addition, the additional efforts to minimize viscous damping was found to be necessary because viscous damping occurring in the channel and around the moving buoy is significant in generation efficiency.

• The Journal of the Korea Contents Association
• /
• v.7 no.11
• /
• pp.111-119
• /
• 2007

Sensor networks are needed for special purposes such as collecting or transmitting information by using sensor devices, for which various routing protocols have been proposed. Among existing protocols, the modified PEGASIS routing technique is known to be effective when applied to cases with directional transmissions of data. However it does not consider recovery from errors or guaranteeing stability in data transmission, while sensor devices performing repetitive relays and controls are prone to errors. In this paper, a double relay routing protocol for Zigbee based sensor networks where data are transfered reliably with a linear direction. The proposed protocol is effective in the sense that it secures the relaible transmission of data with minimal energy consumption based on a directional data transfer. A streetlight control system has been presented as an application of the proposed protocol.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.13 no.4
• /
• pp.395-404
• /
• 2000

A damage detection and assessment algorithm is developed by measuring accelerations at limited locations of a structure under forced vibrations. The developed algorithm applies a time-domain system identification (SI) method that identifies a structure by solving a linearly constrained nonlinear optimization problem for optimal structural parameters. An equation error of the dynamic equilibrium of motion is minimized to estimate optimal parameters. An adaptive parameter grouping scheme is applied to localize damaged members with sparse measured accelerations. Damage is assessed in a statistical manner by applying a time-windowing technique to the measured time history of acceleration. Displacements and velocities at the measured degrees of freedom (DOF) are computed by integrating the measured accelerations. The displacements at the unmeasured DOF are estimated as additional unknowns to the unknown structural parameters, and the corresponding velocities and accelerations we computed by a numerical differentiation. A numerical simulation study with a truss structure is carried out to examine the efficiency of the algorithm. A data perturbation scheme is applied to determine the thresholds lot damage indices and to compute the damage possibility of each member.