• Korean Journal of Food Science and Technology
• /
• v.27 no.4
• /
• pp.525-531
• /
• 1995

Three quality factors of bread during baking process were measured to develop neural network models for bread baking process. Firstly, volume and browning changes during bread baking process were measured using image processing technique and temperature changes inside the bread during process were measured by K-type thermocouples. Relationships among them showed nonlinearity. Secondly, multilayer perception structure with error back propagation learning was used to construct neural network models. Three neural network models for volume, browning, and bread temperature were developed respectively. Developed models showed good performance with predictive error of 4.62% for volume and browning changes after 30 seconds, 7.38% for volume and browning changes after 2 minutes, and 1.09% for temperature change inside the bread respectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.1
• /
• pp.291-299
• /
• 1996

A new identification method using a higher order multilayer neural network is proposed for identifying a complex dynamic system such as a robotic manipulator. The input torque data for learning of the neural identifier are generated for producing effective output trajectories by a minimization process of a specific performance index function which indicates the difference between the reference points and the present joint positions and their velocities of the robotic manipulator. Computer simulation results show that the proposed identification method is very effective for identifying the systems with complex dynamics and large moment of inertia.

• ETRI Journal
• /
• v.23 no.1
• /
• pp.23-32
• /
• 2001

Examination of the statistical variation of integrated passive components is crucial for designing and characterizing the performance of multichip module (MCM) substrates. In this paper, the statistical analysis of parallel plate capacitors with gridded plates manufactured in a multilayer low temperature cofired ceramic (LTCC) process is presented. A set of integrated capacitor structures is fabricated, and their scattering parameters are measured for a range of frequencies from 50 MHz to 5 GHz. Using optimized equivalent circuits obtained from HSPICE, mean and absolute deviation is calculated for each component of each device model. Monte Carlo Analysis for the capacitor structures is then performed using HSPICE. Using a comparison of the Monte Carlo results and measured data, it is determined that even a small number of sample structures, the statistical variation of the component values provides an accurate representation of the overall capacitor performance.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.5 no.1
• /
• pp.11-16
• /
• 2005

This paper presents fabrication process and experimental results of two different types of flexible MEMS biosensors based on polymer/metal multilayer processing techniques. One type of a biosensor is a microelectrode array (MEA) for nerve signal monitoring through implanting the MEA into a living body, and another is a tactile sensor capable of being mounted on an arbitrary-shaped surface. The microelectrode array was fabricated and its electrical characteristics have been examined through in vivo and in vitro experiment. For sensitive skin, flexible tactile sensor array was fabricated and its sensitivity has been analyzed. Mechanical flexibility of these biosensors has been achieved by using a polymer, and it is verified by implanting a MEA to an animal and mounting the tactile sensor on an arbitrary-shaped surface.

• Journal of the Korean Society for Precision Engineering
• /
• v.23 no.11 s.188
• /
• pp.34-41
• /
• 2006

Nanoimprint lithography is an emerging technology which has an ability to make patterns under 100nm width. Recently many researches have been focused to develop multilayer patterning function in nanoimprint lithography and aligning method is attracting attention as a key technology. $Moir\'{e}$ has been used widely to measure dislocation or deformation of objects and considered one of the best solutions to detect aligning error in nanoimprint lithography. Concentric circular patterns are used to generate a $moir\'{e}$ fringe in this paper and aligning offset and direction are extracted from it. Especially this paper shows the difference of fringe equation of $moir\'{e}$ which can be obtained in nanoimprint process atmosphere from normal one.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.644-647
• /
• 2003

This paper describes a neural network based machine vision system designed for inspecting yellow beans in real time. The system consists of a camera. lights, a belt conveyor, air ejectors, and a computer. Beans are conveyed in four lines on a belt and their images are taken by a monochrome line scan camera when they fall down from the belt. Beans are separated easily from their background on images by back-lighting. After analyzing the image, a decision is made by a multilayer artificial neural network (ANN) trained by the error back-propagation (EBP) algorithm. We use the global mean, variance and local change of gray levels of a bean for the input nodes of the network. In an our experiment, the system designed could process about 520kg/hour.

• Proceedings of the Materials Research Society of Korea Conference
• /
• 2011.10a
• /
• pp.28.1-28.1
• /
• 2011

We deposited the carbon films on Ni substrates by thermal atomic layer deposition (th-ALD), for the first time, using carbon tetrabromide ($CBr_4$) precursors and H2 reactants at two different temperatures (573 K and 673 K). Morphology of carbon films was characterized by scanning electron microscopy (SEM). The carbon films having amorphous carbon structures were analyzed by X-ray photoemission spectroscopy (XPS) and Raman spectroscopy. As the working temperature was increased from 573 K to 673 K, the intensity of C1s spectra was increased while that of O1s core spectra was reduced. That is, the purity of carbon films containing bromine (Br) atoms was increased. Also, the thin amorphous carbon films (ALD 3 cycle) were transformed to multilayer graphene segregated on Ni layer, through the post-annealing and cooling process.

• Proceedings of the Materials Research Society of Korea Conference
• /
• 2003.11a
• /
• pp.119-119
• /
• 2003

탄화규소나 열분해 탄소는 고온 특성 및 화학적인 안정성 이 우수하여 단미 혹은 코팅재로로 소재의 성능을 향상시키기 위하여 에너지 관련 분야, 반도체 치구 분야, 방위산업 및 항공우주 분야와 원자력 분야에서 다양하게 사용된다. 특히 원자력 분야에서는 고온형 원자로의 노심 요소 부품으로 적용 및 개발을 고려하고 있으며, 대표적인 예로 수소생산용 초고온 가스냉각로의 코팅 핵연료 입자를 들 수 있다. 일반적으로 TRISO라 불리는 가스냉각로 핵연료는 구형 $UO_2$ kemel의 주변을 PyC-SiC -PyC의 삼중 코팅층으로 둘러싸는 구조를 하고 있으며, 이 코팅층들은 kernel물질이 분열하는 동안 발생되는 내부 기체 압력을 견디는 압력용기 역할과 기체나 금속 핵분열 생성물들을 가두는 확산 장벽 역할을 하게 된다. 본 연구에서는 구형의 $UO_2$대신 선행연구를 위하여 구형 ZrO$_2$를 이용하여 증착온도나 시간 및 입력기체비 등의 화학증착 변수로 조절하여 SiC 및 PyC을 코팅하고, 각 변수들에 의한 증착층의 거동을 고찰하고자 하였다.

• Journal of Korean Institute of Industrial Engineers
• /
• v.25 no.1
• /
• pp.111-124
• /
• 1999

This evapaper is toluate the forecasting performance of three neural network(NN) approaches against ARIMA model using the famous time series analysis competition data. The first NN approach is to analyze the second Makridakis (M2) Competition Data using Multilayer Perceptron (MLP) that has been the most popular NN model in time series analysis. Since it is recently known that MLP suffers from bias/variance dilemma, two approaches are suggested in this study. The second approach adopts Cascade Correlation Network (CCN) that was suggested by Fahlman & Lebiere as an alternative to MLP. In the third approach, a time series is separated into two series using Noise Filtering Network (NFN) that utilizes autoassociative memory function of neural network. The forecasts in the decomposition analysis are the sum of two prediction values obtained from modeling each decomposed series, respectively. Among the three NN approaches, Decomposition Analysis shows the best forecasting performance on the M2 Competition Data, and is expected to be a promising tool in analyzing socio-economic time series data because it reduces the effect of noise or outliers that is an impediment to modeling the time series generating process.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.232-234
• /
• 2017

사이버전 피해평가에 있어서 유무선 통합 환경에 대한 공격의 탐지와 이에 대한 평가가 필요한 상황이다. 특히 회사, 정부 및 군 시설 등에서 인가되지 않은 AP를 사용하여 공격이 발생하는 경우 각종 바이러스 및 해킹 공격에 의한 피해가 발생한 가능성이 높다. 띠라서 인가된 AP와 인가되지 않은 AP를 탐지해서 찾아 내야한다. 본 논문에서는 인가된 AP와 인가 되지 않은 AP를 탐지하기 위해 RTT(Round Trip Time)값을 데이터셋으로 만들고 각 기계학습 알고리즘 SVM(Support Vector Machine), J48(C4.5), KNN(K nearest neighbors), MLP(Multilayer Perceptron)의 결과를 비교해 성능의 차이를 밝히고 이를 통하여 공격을 탐지하여 피해평가에 연결이 되도록 한다.