• 제목/요약/키워드: Remaining Layer

검색결과 211건 처리시간 0.025초

Deep-learning based In-situ Monitoring and Prediction System for the Organic Light Emitting Diode

  • Park, Il-Hoo;Cho, Hyeran;Kim, Gyu-Tae
    • 반도체디스플레이기술학회지
    • /
    • 제19권4호
    • /
    • pp.126-129
    • /
    • 2020
  • We introduce a lifetime assessment technique using deep learning algorithm with complex electrical parameters such as resistivity, permittivity, impedance parameters as integrated indicators for predicting the degradation of the organic molecules. The evaluation system consists of fully automated in-situ measurement system and multiple layer perceptron learning system with five hidden layers and 1011 perceptra in each layer. Prediction accuracies are calculated and compared depending on the physical feature, learning hyperparameters. 62.5% of full time-series data are used for training and its prediction accuracy is estimated as r-square value of 0.99. Remaining 37.5% of the data are used for testing with prediction accuracy of 0.95. With k-fold cross-validation, the stability to the instantaneous changes in the measured data is also improved.

C-MAPSS 데이터를 이용한 항공기 엔진의 신경 회로망 기반 건전성관리 (Neural Network based Aircraft Engine Health Management using C-MAPSS Data)

  • 윤유리;김석구;조성희;최주호
    • 항공우주시스템공학회지
    • /
    • 제13권6호
    • /
    • pp.17-25
    • /
    • 2019
  • 항공기 엔진의 고장예지 및 건전성 관리(PHM)는 고장 또는 수명한계 도달 전에 잔존 유효 수명을 예측하는 것이다. PHM 기술 중 예측모델을 확립하는 방법은 물리 기반과 데이터 기반 방법이 있다. 물리기반 방법은 적은 데이터로 정확한 예측이 가능하지만 확립된 손상 물리 모델이 적어서 적용에 한계가 있다. 본 연구는 따라서 데이터 기반 방법을 적용하였으며, 수명 예측을 위해서 신경회로망 알고리즘 중 Multi-layer Perceptron을 이용하였다. 이를 위해 미국 항공우주국(NASA)에서 개발한 C-MAPSS 코드로 생성된 가상 데이터 세트를 이용하여 신경회로망을 학습하였다. 학습된 신경회로망 모델은 테스트 세트에 적용한 후 잔존 유효 수명의 신뢰구간을 예측하고 실제 값을 통해 정확도를 검증하였다. 또한 본 연구에서 제시된 방법을 기존 문헌의 것과도 비교하였고 그 결과 비교적 양호한 정확도를 확인할 수 있었다.

Ni-assisted growth of transparent and single crystalline indium-tin-oxide nanowires

  • 김현기;김준동;박형호
    • 한국진공학회:학술대회논문집
    • /
    • 한국진공학회 2015년도 제49회 하계 정기학술대회 초록집
    • /
    • pp.259-259
    • /
    • 2015
  • Single crystalline indium-tin-oxide (ITO) nanowires (NWs) were grown by sputtering method. A thin Ni film of 5 nm was deposited before ITO sputtering. Thermal treatment forms Ni nanoparticles, which act as templates to diffuse Ni into the sputtered ITO layer to grow single crystalline ITO NWs. This Ni diffusion through an ITO NW was investigated by transmission electron microscope to observe the Ni-tip sitting on a single crystalline ITO NW. Meanwhile, a single crystalline ITO structure was found at bottom and body part of a single ITO NW without remaining of Ni atoms. This indicates the Ni atoms diffuse through the oxygen vacancies of ITO structure. Rapid thermal process (RTP) applied to generate an initial stage of a formation of Ni nanoparticles with variation in time periods to demonstrate the existence of an optimum condition to initiate ITO NW growth. Modulation in ITO sputtering condition was applied to verify the ITO NW growth or the ITO film growth. The Ni-assisted grown ITO layer has an improved electrical conductivity while maintaining a similar transmittance value to that of a single ITO layer. Electrically conductive and optically transparent nanowire-coated surface morphology would provide a great opportunity for various photoelectric devices.

  • PDF

평판형광램프의 제작 및 특성 (Fabrication and Characteristics of Flat Fluorescent Lamp)

  • 권순석;임민수;임기조
    • 조명전기설비학회논문지
    • /
    • 제16권1호
    • /
    • pp.8-12
    • /
    • 2002
  • 본 논문은 LCD의 백라이트로서 고휘도를 내는 간단한 구조의 평판형광램프를 연구하였다. 평판형광램프는 ITO glass와 절연층, 형광층, 전극층, gas층으로 구성된다. 이 램프의 방전개시전압은 주파수가 증가하면 감소하며 이러한 경향으로는 gas층에서 트립된 Xe과 Ar 양이온에 기인된 공간방전효과의 결과이다. 높은 구동 주파수에서 uniform 전압의 감소는 초기 구동에서 발생된 남아있는 공간전하에 기인한다. 결과로서 Xe gas를 사용한 평판형광램프는 구동전압 700[ $V_{rms}$], 주파수 80[kHz]에서 2700[cd/ $m^2$]의 휘도, 균일도 96[%], 최대 효율은 51[m/W]를 얻었다.다.

비 진공으로 제작한 CIGS 박막 특성 (Characteristics of CIGS film fabricated by non-vacuum process)

  • 박명국;안세진;윤재호;곽지혜;김동환;윤경훈
    • 한국신재생에너지학회:학술대회논문집
    • /
    • 한국신재생에너지학회 2009년도 춘계학술대회 논문집
    • /
    • pp.19-22
    • /
    • 2009
  • A non-vacuum process for fabrication of $CuIn_xGa_{1-x}Se_2$ (CIGS) absorber layer from the corresponing Cu, In, Ga solution precursors was described. Cu, In, Ga precursor solution was prepared by a room temperature colloidal route by reacting the starting materials $Cu(NO_3)_2$, $InCl_3$, $Ga(NO_3)$ and methanol. The Cu, In, Ga precursor solution was mixed with ethylcellulose as organic binder material for the rheology of the mixture to be adjusted for the doctor blade method. After depositing the mixture of Cu, In, Ga solution with binder on Mo/glass substrate, the samples were preheated on the hot plate in air to evaporate remaining solvents and to burn the organic binder material. Subsequently, the resultant CIG/Mo/glass sample was selenized in Se evaporation in order to get a solar cell applicable dense CIGS absorber layer. The CIGS absorber layer selenized at $530^{\circ}C$ substrate temperature for 1h with various metal organic ratio.

  • PDF

상변화 물질의 용융과정에 있어서 좌표변환을 이용한 온도분포의 해석적 연구 (The finite difference analysis on temperature distribution by coordinate transformation during melting process of phase-change Material)

  • 김준근;임장순
    • 태양에너지
    • /
    • 제5권2호
    • /
    • pp.77-83
    • /
    • 1985
  • An analysis is performed to investigate the influence of the buoyancy force and the thickness variation of melting layer in the containment that is filled with phase-change Material surrounding a cylindrical heating tube during melting process. The phase-change material is assumed to be initially solid at its phase-change temperature and the remaining solid at any given time is still at the phase-change temperature and neglecting the effect of heat transfer occuring within the solid. At the start of melting process, the thickness of melting layer is assumed to be a stefan-problem and after the starting process, the change of temperature and velocity is calculated using a two dimensional finite difference method. The governing equations for velocity and temperature are solved by a finite difference method which used SIMPLE (Semi Implicit Method Pressure linked Equations) algorithm. Results are presented for a wide range of Granshof number and in accordance with the time increment and it is founded that two dimensional fluid flow occurred by natural convection decreases the velocity of melting process at the bottom of container. The larger the radius of heating tube, the higher heat transfer is occurred in the melting layer.

  • PDF

K-FILE과 초음파 기구의 도말층제거효과에 대한 주사전자 현미경적 연구 (A SCANNING ELECTRON MICROSCOPIC STUDY ON THE REMOVING EFFICIENCY OF SMEAR LAYER BY K-FILE AND ULTRASONIC INSTRUMENT)

  • 이수종;임미경
    • Restorative Dentistry and Endodontics
    • /
    • 제19권1호
    • /
    • pp.97-105
    • /
    • 1994
  • The purpose of this study was to evaluate the smear layer removing efficiency of two root canal preparation techniques. Twelve single-rooted teeth were used in two groups of six each. Group 1 was biomechanically prepared by hand using a K-file with a high volume of normal saline irrigation. Group 2 was. prepared by using ultrasonically activated K-file with a constant high volume of normal saline irrigation. After the experimental procedures, each root was split saggitally. The removing efficiency of the preparation methods were assessed in terms of surface condition of the canal walls at three levels, those coronal, middle, and apical thirds. On the basis of remaining debris, presence of smear layer, and patency of dentinal tubules, each canal was evaluated according to a scale form 0 to 2. A statistical analysis was used to indicated any significant differences in surface condition between the two methods. There was no statistical significance between hand instrumentation and ultrasonic instrumentation at the cervical third but removing efficiency of ultrasonic instrumentation was superior. No statistically significant differences were obhserved for middle or apical third.

  • PDF

비진공 나노입자 코팅법을 이용한 CIGS 박막 태양전지 제조 (Fabrication of CIGS Thin Film Solar Cell by Non-Vacuum Nanoparticle Deposition Technique)

  • 안세진;김기현;윤경훈
    • 한국신재생에너지학회:학술대회논문집
    • /
    • 한국신재생에너지학회 2006년도 춘계학술대회
    • /
    • pp.222-224
    • /
    • 2006
  • A non-vacuum process for $Cu(In,Ga)Se_2$ (CIGS) thin film solar cells from nanoparticle precursors was described in this work CIGS nanoparticle precursors was prepared by a low temperature colloidal route by reacting the starting materials $(CuI,\;InI_3,\;GaI_3\;and\;Na_2Se)$ in organic solvents, by which fine CIGS nanoparticles of about 20nm in diameter were obtained. The nanoparticle precursors were mixed with organic binder material for the rheology of the mixture to be adjusted for the doctor blade method. After depositing the mixture of CIGS with binder on Mo/glass substrate, the samples were preheated on the hot plate in air to evaporate remaining solvents ud to burn the organic binder material. Subsequently, the resultant (porous) CIGS/Mo/glass simple was selenized in a two-zone Rapid Thermal Process (RTP) furnace in order to get a solar ceil applicable dense CIGS absorber layer. Complete solar cell structure was obtained by depositing. The other layers including CdS buffer layer, ZnO window layer and Al electrodes by conventional methods. The resultant solar cell showed a conversion efficiency of 0.5%.

  • PDF

고충격에 신관의 생존성을 향상시키기 위한 다층 충격완충장치 전산해석 연구 (Numerical Investigation for Multi-layer Shock Absorber to Improve Survivability of Fuze at High Impact)

  • 소경재;김민겸;이대희
    • 한국전산구조공학회논문집
    • /
    • 제33권4호
    • /
    • pp.255-261
    • /
    • 2020
  • 본 논문은 충격을 줄이기 위해 효과적인 충격완충장치를 구성하는 방법을 제안했다. 기존의 충격완충장치는 폴리에틸렌으로만 만들어졌지만, 새로운 충격완충장치는 외측에는 폴리에틸렌, 내측에는 고밀도 재료로 구성하였다. 충격은 내측과 외측 물질 사이의 밀도 차이가 더 클 때 줄어들었다. 2층 구조의 외측으로 설계하기 위해 알루미늄, 티타늄, 구리를 선택하였다. 가장 밀도가 높은 구리에서는 충격 감소가 가장 좋았으며, 기존 충격완충장치보다 최대 감가속도는 43%, 충격량은 51% 감소하였다. 4층, 6층 충격완충장치의 경우, 충격량은 줄였지만, 최대 감가속도는 증가하였다. 신관은 가장 큰 충격으로부터 살아남아야 하며 나머지 충격파는 임계값을 초과하지 않으므로, 본 논문은 폴리에틸렌-구리를 사용한 2층 구조용 충격완충장치를 제안하였다.

Ag 도금 Cu 입자의 제조에서 암모늄 기반 혼합 용매를 사용한 Cu 입자의 전처리 조건과 이의 영향 (Pretreatment Condition of Cu by Ammonium-Based Mixed Solvent and Its Effects on the Fabrication of Ag-Coated Cu Particles)

  • 이희범;이종현
    • 한국재료학회지
    • /
    • 제26권3호
    • /
    • pp.109-116
    • /
    • 2016
  • To achieve the fabrication of high-quality Ag-coated Cu particles through a wet chemical process, we reported herein pretreatment conditions using an ammonium-based mixed solvent for the removal of a $Cu_2O$ layer on Cu particles that were oxidized in air for 1 hr at $200^{\circ}C$ or for 3 days at room temperature. Furthermore, we discussed the results of post-Ag plating with respect to removal level of the oxide layer. X-ray diffraction results revealed that the removal rate of the oxide layer is directly proportional to the concentration of the pretreatment solvent. With the results of Auger electron spectroscopy using oxidized Cu plates, the concentrations required to completely remove 50-nm-thick and 2-nm-thick oxides within 5 min were determined to be X2.5 and X0.13. However, the optimal concentrations in an actual Ag plating process using Cu powder increased to X0.4 and X0.5, respectively, because the oxidation in powder may be accelerated and the complete removal of oxide should be tuned to the thickest oxide layer among all the particles. Back-scattered electron images showed the formation of pure fine Ag particles instead of a uniform and smooth Ag coating in the Ag plating performed after incomplete removal of the oxide layer, indicating that the remaining oxide layer obstructs heterogeneous nucleation and plating by reduced Ag atoms.