• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제1권1호
• /
• pp.87-94
• /
• 2001

Silicon nitride films grown by plasma-enhanced chemical vapor deposition (PECVD) are useful for a variety of applications, including anti-reflecting coatings in solar cells, passivation layers, dielectric layers in metal/insulator structures, and diffusion masks. PECVD systems are controlled by many operating variables, including RF power, pressure, gas flow rate, reactant composition, and substrate temperature. The wide variety of processing conditions, as well as the complex nature of particle dynamics within a plasma, makes tailoring SiN film properties very challenging, since it is difficult to determine the exact relationship between desired film properties and controllable deposition conditions. In this study, SiN PECVD modeling using optimized neural networks has been investigated. The deposition of SiN was characterized via a central composite experimental design, and data from this experiment was used to train and optimize feed-forward neural networks using the back-propagation algorithm. From these neural process models, the effect of deposition conditions on film properties has been studied. A recipe synthesis (optimization) procedure was then performed using the optimized neural network models to generate the necessary deposition conditions to obtain several novel film qualities including high charge density and long lifetime. This optimization procedure utilized genetic algorithms, hybrid combinations of genetic algorithm and Powells algorithm, and hybrid combinations of genetic algorithm and simplex algorithm. Recipes predicted by these techniques were verified by experiment, and the performance of each optimization method are compared. It was found that the hybrid combinations of genetic algorithm and simplex algorithm generated recipes produced films of superior quality.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 C
• /
• pp.1180-1182
• /
• 1999

Modeling for utility interactive photovoltaic power generation system has been studied using PSCAD/EMTDC. The proposed model system consists of a simple utility circuit configuration, 3kW of single phase utility interactive photovoltaic system, single phase PWM voltage source inverter module, and feed forward PID controller as control circuit. In the system, the DC current is assumed constant, and the voltage source inverter provides sinusoidal ac current for the loads of utility system. The simulation results are given in order to verify the effectiveness of the proposed model. The phases of output voltage of utility system and the output current of the inverter module are compared. Especially, the compensation effect of the photovoltaic system for the unbalanced load is analyzed. and the transient phenomena for a phase to ground fault are also simulated.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 1995년도 추계학술대회논문집; 한국종합전시장, 24 Nov. 1995
• /
• pp.261-266
• /
• 1995

본 연구에서는 자기부상시스템에 대해 흡인식 자기부상방식을 채택하고 쇠구슬에 대한 운동을 상하 1자유도로 가정하여 운동방정식을 세운다. 이때 전자석이 자기 부상력은 전자석에 흐르는 전류와 인덕턴스의 함수라 가정하고, 모델의 불확실성은 자기부상계의 운동 방정식으로부터 선형화 할 때 발생하는 오차 및 파라미터 변동으로 생각한다. 또한 모델의 불확실성이 존재하더라도 정상편차 없이 부상하는 서보제어계를 설계한다. 그런데 저자등은 강인성 문제 및 정상편차 없는 것에 역점을 두어 H$_{\infty}$ 제어이론에 기초한 1형 로바스트 서보 제어기를 구하여 자기부상 시스템의 안정화 제어계로써 적용한 적이 있다. 이때 중심해 이외의 해를 이용하여 설계한 서보 제어계는 자기부상계의 과도상태시에 일어나는 오버슈트를 줄일 수 없었다. 따라서 시스템 내부 안정화를 위하여 H$_{\infty}$ 제어이론에 의해 설계된 피드백(feedback) 제어기와 물체가 부상할 때 오버슈트를 줄이고 제어량이 목표치에 잘 추종하기 위해 설계된 피드 포워드(feed forward) 제어기로써 2자유도를 갖는 제어계를 설계한다. 이렇게 설계한 2자유도 제어계를 가지고 모의 응답실험과 본 연구자들이 만든 자기부상 시스템의 실험결과를 비교함으로써 설계된 제어기의 타당성을 조사한다.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2007년도 춘계종합학술대회
• /
• pp.171-174
• /
• 2007

본 논문은 역전파 알고리즘을 이용한 FF-PID 제어 시스템 구현에 관한 연구이다. 영상의 인식은 신경망 역전파를 사용하여 학습시킨다. FF-PID 제어기는 신경망의 목표치에 대한 출력층 오차값을 제어값으로 사용하여 이동물체의 응답특성을 향상시킨다. 실험결과, 시스템의 응답시간은 약 2.7(sec)였으며, 일반적인 차영상기법에 비하여 약 15% 목표치 응답이 향상되어, 효과적인 이동물체의 제어가 가능하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 추계학술대회 논문집 전기기기 및 에너지변환시스템부문
• /
• pp.160-163
• /
• 2004

A passive lossless turn-on/turn-off snubber network is proposed for the boost PWM converter. Previous AC/DC PFC Boost Converter perceives feed forward signal of output for average current-mode control. Previous Boost Convertor, the Quantity of input current will be decreased by the decrease of output current in light load, and also Power factor comes to be decreased. Also the efficiency of converter will be decreased by the decrease of power factor. The proposed converter presents the good PFC, low line current harmonic distortions and tight output voltage regulations using energy recovery circuit. All of the semiconductor devices in the converter are turned on under exact or near zero voltage switching(ZVS). No additional voltage and current stresses on the main switch and main diode occur. To show the superiority of this converter is verified through the experiment with a 640W, 100kHz prototype converter.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1998년도 제13차 학술회의논문집
• /
• pp.195-200
• /
• 1998

Since the neural network was introduced, significant progress has been made on data handling and learning algorithms. Currently, the most popular learning algorithm in neural network training is feed forward error back-propagation (FFEBP) algorithm. Aside from the success of the FFEBP algorithm, polynomial neural networks (PNN) learning has been proposed as a new learning method. The PNN learning is a self-organizing process designed to determine an appropriate set of Ivakhnenko polynomials that allow the activation of many neurons to achieve a desired state of activation that mimics a given set of sampled patterns. These neurons are interconnected in such a way that the knowledge is stored in Ivakhnenko coefficients. In this paper, the PNN model has been developed using the plasma enhanced chemical vapor deposition (PECVD) experimental data. To characterize the PECVD process using PNN, SiO$_2$films deposited under varying conditions were analyzed using fractional factorial experimental design with three center points. Parameters varied in these experiments included substrate temperature, pressure, RF power, silane flow rate and nitrous oxide flow rate. Approximately five microns of SiO$_2$were deposited on (100) silicon wafers in a Plasma-Therm 700 series PECVD system at 13.56 MHz.

• 조명전기설비학회논문지
• /
• 제18권6호
• /
• pp.128-135
• /
• 2004

본 논문은 자기부상열차용 전원의 문제점을 개선시키기 위해서 다중루프 제어기를 제시하였다. 제시된 제어기는 3개의 부분으로 구성되어 있다. 첫 번째는 입력전압의 변동에 대하여 보상할 수 있는 Feed Forward제어기이며 두 번째는 리액터 전류와 출력 전류의 차를 보상하며, 세 번째는 비례적분제어기를 사용하여 출력전압에 포함된 리플을 감소시키므로써, 안정화된 시스템을 구현하였다. 이 시스템의 특성을 확인하기 위해서 Matlab Simulink와 고성능 DSP소자인 TMS320F240을 이용하여 비교 분석하였다.

• ETRI Journal
• /
• 제41권2호
• /
• pp.235-241
• /
• 2019

In this paper, we propose a rank-weighted reconstruction feature to improve the robustness of a feed-forward deep neural network (FFDNN)-based acoustic model. In the FFDNN-based acoustic model, an input feature is constructed by vectorizing a submatrix that is created by slicing the feature vectors of frames within a context window. In this type of feature construction, the appropriate context window size is important because it determines the amount of trivial or discriminative information, such as redundancy, or temporal context of the input features. However, we ascertained whether a single parameter is sufficiently able to control the quantity of information. Therefore, we investigated the input feature construction from the perspectives of rank and nullity, and proposed a rank-weighted reconstruction feature herein, that allows for the retention of speech information components and the reduction in trivial components. The proposed method was evaluated in the TIMIT phone recognition and Wall Street Journal (WSJ) domains. The proposed method reduced the phone error rate of the TIMIT domain from 18.4% to 18.0%, and the word error rate of the WSJ domain from 4.70% to 4.43%.

• Environmental Engineering Research
• /
• 제24권3호
• /
• pp.404-411
• /
• 2019

In this study, we designed a data-driven model to predict chlorophyll-a using M5P model tree and extreme learning machine (ELM). The Juksan weir in the Youngsan River has high chlorophyll-a, which is the primary indicator of algal bloom every year. Short-term algal bloom prediction is important for environmental management and ecological assessment. Two models were developed and evaluated for short-term algal bloom prediction. M5P is a classification and regression-analysis-based method, and ELM is a feed-forward neural network with fast learning using the least square estimate for regression. The dataset used in this study includes water temperature, rainfall, solar radiation, total nitrogen, total phosphorus, N/P ratio, and chlorophyll-a, which were collected on a daily basis from January 2013 to December 2016. The M5P model showed that the prediction model after one day had the highest performance power and dropped off rapidly starting with predictions after three days. Comparing the performance power of the ELM model with the M5P model, it was found that the performance power of the 1-7 d chlorophyll-a prediction model was higher. Moreover, in a period of rapidly increasing algal blooms, the ELM model showed higher accuracy than the M5P model.

• 대한전자공학회논문지SD
• /
• 제46권3호
• /
• pp.51-59
• /
• 2009

본 논문에서는 어쿠스틱 센서 IC 용 연속 시간 시그마-델타 모듈레이터를 구현하였다. 모듈레이터의 전력 소모를 최소화하기 위해 summing 단의 필요성을 제거한 피드-포워드 (feed-forward) 구조로 설계 하였으며, 해상도를 높이기 위해 선형성이 우수한 active-RC 필터를 사용하여 설계 하였다. 또한 초과 루프 지연 시간 (excess loop delay)에 의한 성능 저하를 방지하기 위한 회로 기법을 제안 하였다. 저 전압, 고 해상도의 4차 단일 비트 연속 시간 시그마-델타 모듈레이터는 $0.13{\mu}m$ 1 poly 8 metal CMOS 표준 공정으로 제작하였으며 코어 크기는 $0.58\;mm^2$ 이다 시뮬레이션 결과 25 kHz 의 신호 대역 내에서 91.3 dB의 SNR(signal to noise ratio)을 얻었고 전체 전력 소모는 $290{\mu}W$ 임을 확인하였다.