• Management Science and Financial Engineering
• /
• v.17 no.1
• /
• pp.19-43
• /
• 2011

Tuning the architecture of SVM (support vector machine) is to build an SVM model of better performance. Two different tuning methods of the grid search and the GA (genetic algorithm) have been addressed in the literature, each of which has its own methodological pros and cons. This paper suggests a combined method for tuning the architecture of SVM models, which employs the GAM (generalized additive models), the grid search, and the GA in sequence. The GAM is used for selecting input variables, and the grid search and the GA are employed for finding optimal parameter values of the SVM models. Applying the method to a bankruptcy prediction problem, we show that SVM model tuned by the proposed method outperforms other SVM models.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.26 no.5
• /
• pp.325-333
• /
• 2021

A 3 kW grid-tied PV inverter with Gallium nitride high-electron mobility transistor (GaN HEMT) for domestic commercialization was developed using boost converter and full-bridge inverter with LCL filter topology. Recently, many GaN HEMTs are manufactured as surface mount packages because of their lower parasitic inductance characteristic than standard TO (transistor outline) packages. A surface mount packaged GaN HEMT releases heat through either top or bottom cooling method. IGOT60R070D1 is selected as a key power semiconductor because it has a top cooling method and fairly low thermal resistances from junction to ambient. Its characteristics allow the design of a 3 kW inverter without forced convection, thereby providing great advantages in terms of easy maintenance and high reliability. 1EDF5673K is selected as a gate driver because its driving current and negative voltage output characteristics are highly optimized for IGOT60R070D1. An LCL filter with passive damping resistor is applied to attenuate the switching frequency harmonics to the grid-tied operation. The designed LCL filter parameters are validated with PSIM simulation. A prototype of 3 kW PV inverter with GaN HEMT is constructed to verify the performance of the power conversion system. It achieved high power density of 614 W/L and peak power efficiency of 99% for the boost converter and inverter.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2013.02a
• /
• pp.411-412
• /
• 2013

최근 UV LED는 생화학 및 의료 산업에서 많은 각광을 받고 있다. 특히, 360nm 이하의 파장대를 갖는 UV LED는 치료 기술, 센서, 물이나 공기 등의 정화와 같은 목적으로 특별한 관심이 쏠리고 있다 [1]. 이러한 지속적인 연구를 통하여 현재까지 UV LED는 거대한 성장을 이루어 왔다. 하지만 이러한 노력에도 불구하고, 360 nm 이하의 UV LED는 여전히 오믹 접촉과 전류 분산이 원활하지 못하다는 문제점을 가지고 있다. 이것은 UV LED의 외부 양자 효율을 감소시키고, 더 나아가 극도로 낮은 광 추출 효율을 초래한다. 최근 이러한 문제를 해결하고자, 투명 전도성 산화물(TCO)을 금속 전극과 p-AlGaN 사이에 삽입해주는데, 현재 가장 널리 사용되는 TCO 물질은 ITO 이다 [2]. 하지만 ITO 물질은 상대적으로 작은 밴드갭(3.3~4.3 eV)과 단파장 빛이 가지는 큰 에너지로 인하여 deep-UV 영역에서는 빛이 투과하지 못하고 대부분 흡수된다 [3]. 따라서 본 연구에서는 기존의 박막형 ITO 투명 전극에 비해 투과도 손실을 최소화할 수 있는 mesh, grid 기반의 투명전극을 연구하였다. Fig. 1과 같이 $5{\mu}m$, $10{\mu}m$, $20{\mu}m$ 간격으로 이루어진 mesh, grid 구조의 투명전극을 구현하여 투과도 손실을 최소화하면서 우수한 전기적 특성을 확보하기 위한 구조 최적화 연구를 진행하였다. 본 연구를 위해 mesh, grid 구조의 ITO 전극 패턴을 photolitho 공정으로 형성하였으며, e-beam 증착법으로 60 nm 두께의 ITO 전극을 형성 후 질소 분위기/$650^{\circ}$에서 30초 동안 RTA 공정을 진행하였다. Fig. 1에서 볼 수 있듯이 mesh, grid의 간격이 증가할수록 투명 전극이 차지하는 면적이 감소하여 투과도는 향상되는 반면, 투명 전극과 p-GaN과의 접촉 면적 또한 감소하므로 오믹 특성이 저하된다. 따라서 투과도 손실을 최소화하면서 우수한 전기적 특성을 확보하기 위해 mesh는 $20{\mu}m$, grid는 $10{\mu}m$ 간격의 구조로 각각 최적화하였다. 그 결과 박막 기반의 ITO 투명전극 대비 최대 약 10% 향상된 투과도를 확보하였으며, I-V Curve 결과를 통하여 p-GaN 기판과 mesh 구조의 ITO 전극 사이에 박막 기반의 투명 전극과 비슷한 수준인 $0.35{\mu}A(@5V)$의 전기적 특성을 확보하였다. 결과적으로 mesh, grid 기반 투명전극의 구조 최적화를 통하여 p-GaN과 원활한 오믹 접촉을 형성하는 동시에 기존 박막형 ITO 투명 전극 구조보다 높은 투과도를 확보할 수 있었다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2001.01a
• /
• pp.37-44
• /
• 2001

This paper describes the modeling of work environment for the extraction of abstract operation rules for cooperative work with multiple agent. We propose the modeling method using a potential field. In the method, it is applied to a box pushing problem, which is to move a box from a start to a goal b multiple agent. The agents follow the potential value when they move and work in the work environment. The work environment is represented as the grid space. The potential field is generated by Genetic Algorithm(GA) for each agent. GA explores the positions of a potential peak value in the grid space, and then the potential value stretching in the grid space is spread by a potential diffusion function in each grid. However it is difficult to explore suitable setting using hand coding of the position of peak potential value. Thus, we use an evlolutionary computation way because it is possible to explore the large search space. So we make experiments the environment modeling using the proposed method and verify the performance of the exploration by GA. And we classify some types from acquired the environment model and extract the abstract operation rule, As results, we find out some types of the environment models and operation rules by the observation, and the performance of GA exploration is almost same as the hand coding set because these are nearly same performance on the evaluation of the consumption of agent's energy and the work step from point to the goal point.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.1
• /
• pp.64-71
• /
• 2017

Considering the volatility, intermittent and random of photovoltaic (PV) generation systems, accurate forecasting of PV power output is important for the grid scheduling and energy management. In order to improve the accuracy of short-term power forecasting of PV systems, this paper proposes a prediction model based on environmental factors and support vector machine optimized by genetic algorithm (GA-SVM). In order to improve the prediction accuracy of this model, weather conditions are divided into three types, and the gray correlation coefficient algorithm is used to find out a similar day of the predicted day. To avoid parameters optimization into local optima, this paper uses genetic algorithm to optimize SVM parameters. Example verification shows that the prediction accuracy in three types of weather will remain at between 10% -15% and the short-term PV power forecasting model proposed is effective and promising.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.606-609
• /
• 2015

본 연구에서는 support vector regression (SVR) 및 매개변수 최적화 알고리즘을 이용한 하천수위 예측모델을 구축하고 이를 실제 유역에 적용하여 모델 효율성을 평가하였다. 여기서, SVR은 하천수위를 예측하기 위한 예측모델로서 채택되었으며, 커널함수 (Kernel function)로서는 radial basis function (RBF)을 선택하였다. 최적화 알고리즘은 SVR의 최적 매개변수 (C?, cost parameter or regularization parameter; ${\gamma}$, RBF parameter; ${\epsilon}$, insensitive loss function parameter)를 탐색하기 위하여 적용되었다. 매개변수 최적화 알고리즘으로는 grid search (GS), genetic algorithm (GA), particle swarm optimization (PSO), artificial bee colony (ABC) 알고리즘을 채택하였으며, 비교분석을 통해 최적화 알고리즘의 적용성을 평가하였다. 또한 SVR과 최적화 알고리즘을 결합한 모델 (SVR-GS, SVR-GA, SVR-PSO, SVR-ABC)은 기존에 수자원 분야에서 널리 적용되어온 신경망(Artificial neural network, ANN) 및 뉴로퍼지 (Adaptive neuro-fuzzy inference system, ANFIS) 모델과 비교하였다. 그 결과, 모델 효율성 측면에서 SVR-GS, SVR-GA, SVR-PSO 및 SVR-ABC는 ANN보다 우수한 결과를 나타내었으며, ANFIS와는 비슷한 결과를 나타내었다. 또한 SVR-GA, SVR-PSO 및 SVR-ABC는 SVR-GS보다 상대적으로 우수한 결과를 나타내었으며, 모델 효율성 측면에서 SVR-PSO 및 SVR-ABC는 가장 우수한 모델 성능을 나타내었다. 따라서 본 연구에서 적용한 매개변수 최적화 알고리즘은 SVR의 매개변수를 최적화하는데 효과적임을 확인할 수 있었다. SVR과 최적화 알고리즘을 이용한 하천수위 예측모델은 기존의 ANN 및 ANFIS 모델과 더불어 하천수위 예측을 위한 효과적인 도구로 사용될 수 있을 것으로 판단된다.

• 한국신재생에너지학회:학술대회논문집
• /
• 2006.06a
• /
• pp.199-201
• /
• 2006

A lot of researches and experiments have been performed to make more efficient dye-sensitized solar cell. Among them, insertion of a grid electrode in DSC is one method to increase overall performance as being shortened distance of electrons' movement by diffusion. In this paper, we measured overall characteristics which is included voltage-current characteristic curve, efficiency, fill factor by comparison between DSC without and with grid electrode. As a result, we got maximum 1.8 times increase of efficiency and approximately 1.7 times rise of fill factor. And we experimented 8, 10, 12, 14mm of grid gap and compared their results to optimize grid gap which is able to incite the most effective movement of electrons. Consequently, we got the result that electrons' movement is the most effect ive when grid gap is approximately 12mm.

• Applied Microscopy
• /
• v.50
• /
• pp.22.1-22.6
• /
• 2020

In-situ transmission electron microscopy (TEM) holders that employ a chip-type specimen stage have been widely utilized in recent years. The specimen on the microelectromechanical system (MEMS)-based chip is commonly prepared by focused ion beam (FIB) milling and ex-situ lift-out (EXLO). However, the FIB-milled thin-foil specimens are inevitably contaminated with Ga⁺ ions. When these specimens are heated for real time observation, the Ga⁺ ions influence the reaction or aggregate in the protection layer. An effective method of removing the Ga residue by Ar⁺ ion milling within FIB system was explored in this study. However, the Ga residue remained in the thin-foil specimen that was extracted by EXLO from the trench after the conduct of Ar⁺ ion milling. To address this drawback, the thin-foil specimen was attached to an FIB lift-out grid, subjected to Ar⁺ ion milling, and subsequently transferred to an MEMS-based chip by EXLO. The removal of the Ga residue was confirmed by energy dispersive spectroscopy.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.19 no.2
• /
• pp.161-176
• /
• 2021

In this study, the electrochemical behavior of Sm on the binary liquid Al-Ga cathode in the LiCl-KCl molten salt system is investigated. First, the co-reduction process of Sm(III)-Al(III), Sm(III)-Ga(III), and Sm(III)-Ga(III)-Al(III) on the W electrode (inert) were studied using cyclic voltammetry (CV), square-wave voltammetry (SWV) and open circuit potential (OCP) methods, respectively. It was identified that Sm(III) can be co-reduced with Al(III) or Ga(III) to form Al_zSm_y or Ga_xSm_y intermetallic compounds. Subsequently, the under-potential deposition of Sm(III) at the Al, Ga, and Al-Ga active cathode was performed to confirm the formation of Sm-based intermetallic compounds. The X-ray diffraction (XRD) and scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS) analyses indicated that Ga₃Sm and Ga₆Sm intermetallic compounds were formed on the Mo grid electrode (inert) during the potentiostatic electrolysis in LiCl-KCl-SmCl₃-AlCl₃-GaCl₃ melt, while only Ga₆Sm intermetallic compound was generated on the Al-Ga alloy electrode during the galvanostatic electrolysis in LiCl-KCl-SmCl₃ melt. The electrolysis results revealed that the interaction between Sm and Ga was predominant in the Al-Ga alloy electrode, with Al only acting as an additive to lower the melting point.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.8 no.3
• /
• pp.23-33
• /
• 2005

Hydrologic-topographical parameters were extracted using GIS. The use of GIS is more effective and exact than the execution by person. And the purpose of this paper is to extract more efficient size of grid for DEM analysis by applying GIS technology. As a result, when the grid size is less than $100m{\times}100m$ the trend of extracted parameters is similar but when the grid size is over $100m{\times}100m$ the trend of extraction parameters is dispersive. Therefore, it is appropriate to extract hydrologic-topographical parameters the grid size of $100m{\times}100m$ in DEM analysis.