• 전기학회논문지
• /
• 제61권3호
• /
• pp.407-412
• /
• 2012

This study investigated the process of thermal-induced growth of micro-crack developed at the crystalline solar cell using EL image, determined the output characteristic according to the pattern of micro-crack, analyzed the I-V characteristic according to the pattern of crack growth, and predicted the output value using simulation. The purpose of this study was, therefore, to investigate the process of thermal-induced growth of micro-crack developed at the early stage of PV module completion using EL image, to analyze the resulting decrement of output and predict the output value using simulation. It was observed that the crack grew increasingly by the thermal condition, and accordingly the lowering of output was accelerated. The output values of crack patterns with various direction were predicted using simulation, resulting in close I-V curve with only around 4% of error rate. It is considered that it is possible to predict the electric characteristic of solar cell module using only pattern of micro-crack occurred at solar cell based on our results.

• 한국통신학회논문지
• /
• 제16권2호
• /
• pp.128-136
• /
• 1991

A neural network of multi layer perception type, learned by error back propagation learning rule, is generally used for the verification or clustering of similar type of patterns. When learning is completed, the network has a constant value of output depending on a pattern. This paper shows that the intensity of neuron's out put can be controlled by a function which intensifies the excitatory interconnection coefficients or the inhibitory one between neurons in output layer and those in hidden layer. In this paper the value of factor in the function to control the output is derived from the know values of the neural network after learning is completed And also this paper show that the amount of an increased neuron's output in output layer by arbitary value of the factor is derived. For the applications increased recognition performance of a pattern than has distortion is introduced and the output of partially damaged neurons are first managed and this paper shows that the reduced recognition performance can be recovered.

• 전기학회논문지
• /
• 제60권10호
• /
• pp.1823-1831
• /
• 2011

When the RPS(Renewable Portfolio Standards) becomes effective in 2012, the use of renewable energy will be dramatically increased. However, there are no production simulations and demand supply programs that reflect the characteristics of the renewable energy. This paper analyzes correlations of the domestic wind power and solar power generation pattern in different areas and those of these sources' output and load pattern. Based on the regional correlation analysis, an appropriate method that uses a average output of the renewable energy or another modeling that takes account of uncertainty could be selected. Because it's output is dependent on weather condition, we can not control the generation of renewable energy, that is the reason why the correlation between the load and output pattern of sources can be helpful to determine whether the renewable energy is modeled as a generator or load modifier. Through this analysis, a basis will be provided in order to properly model the renewable energy source.

• 생물환경조절학회지
• /
• 제20권2호
• /
• pp.63-71
• /
• 2011

영상처리는 정확한 오이의 형상 및 위치를 인식하기 위하여 형상인식 알고리즘에 대한 연구를 수행하였다. 다양한 오이형상을 인식하기 위한 방법으로는 신경회로망의 연상 메모리 알고리즘을 이용하여 오이의 특정형상을 인식하였다. 형상인식은 실제영상에서 오이의 형상과 위치를 판정할 수 있도록 알고리즘을 개발한 결과, 다음과 같은 결론을 얻었다. 본 알고리즘에서는 일정한 학습패턴의 수를 2개, 3개, 4개를 각각 기억시켜 샘플패턴 20개를 실험하여 연상시킨 결과, 학습패턴으로 복원된 출력패턴의 비율은 각각 65.0%, 45.0%, 12.5%로 나타났다. 이는 학습패턴의 수가 많을수록 수렴할 때, 다른 출력패턴으로 많이 검출되었다. 오이의 특정형상 검출은 $30{\times}30$간격으로 자동검출 되도록 처리하였다. 실제영상에서 자동 검출로 처리한 결과, 오이인식의 처리시간은 약 0.5~1초/1개(패턴) 빠르게 검출되었다. 또한, 다섯 개의 실제 영상에서 실험한 결과, 학습패턴에 대한 다른 출력패턴은 96~99%의 제거율을 나타내었다. 오이로 인식된 출력패턴 중에서, 오검출된 출력패턴의 비율은 0.1~4.2%를 나타내었다. 본 연구에서는 신경회로망을 이용하여 오이의 형상 및 위치를 인식할 수 있도록 알고리즘을 개발하였다. 오이의 위치측정은 실제영상에서 학습패턴과 유사한 출력패턴의 좌표를 가지고, 오이의 위치좌표를 추정할 수 있었다.

• 대한전자공학회논문지SD
• /
• 제48권6호
• /
• pp.39-44
• /
• 2011

본 논문은 배터리로부터 입력받은 제한된 전압 값을 변환하여 PMIC의 출력 전압 값에 대한 PCB 패턴의 두 가지 모드로 설계한 PCB 패턴의 특성분석을 제안하고자 한다. PCB 설계 기술은 EMI/EMC, 크로스토크, 임피던스 증가 등으로 국내에서는 관련 기술의 미확보로 많은 어려움을 겪고 있으며, 사용되는 용도에 따라 패턴 상호간의 적절한 이격거리 확보, 전류량에 따른 PCB 패턴 폭 등의 기술이 요구되고 있다. 본 논문에서는 PMIC의 출력 전압 값을 커패시터를 거치지 않고 직접 출력하는 방법[모드1]과 커패시터를 거쳐 출력하는 방법[모드2]으로 설계하여 PCB 패턴의 특성 분석을 수행하였다. 또한 IPC-2221에서 제시한 식을 이용하여 전류량에 따른 패턴 폭을 계산해 보았으며, 실험을 통한 패턴 설계에 따른 문제점을 분석하여 올바른 설계 방향을 제시하였다. 이러한 연구는 비단 모바일 패널 전원 공급을 위한 PMIC 설계에 적용될 뿐만 아니라 자동차, 카메라, 노트북, PDA 등의 다양한 분야의 전력 반도체 개발에 적용될 수 있을 것으로 기대된다.

• 정보보호학회논문지
• /
• 제9권4호
• /
• pp.37-48
• /
• 1999

현대암호방식은 종래의 선형 대수와 수리이론을 적용한 암호통신을 벗어난 유사 잡음성을 띠는 카오스 신호를 이용한 암호통신을 적용해 오고 있다,[1-2] 본 논문은 1차 permutation 알고 리즘을 이용 하여 변환된 정보를 2차 이산 카오스 변환 함수를 이용해 암호화하는 광영상 암호시스템을 제안하여UT 다. 제안된 시스템은 키수열 발생기의 출력을 통해 영상정보를 permutation 하는 알고리즘 을 설계하였고 이에 대한 검정을 수행하였다. 또한 본 논문에서는 permutation 알고리즘을 통해 제한적인 카오스 함수 의 적용시 발생하는 문제점을 해결하고 비도를 증가시킴으로써 광영상 암호시스템에 적용 시 그 타당성 을 검정하였다. Current encryption methods have been applied to secure communication using discrete chaotic system whose output is a noise-like signal which differs from the conventional encryption methods that employ algebra and number theory[1-2] We propose an optical encryption method that transforms the primary pattern into the image pattern of discrete chaotic function first a primary pattern is encoded using permutation algorithm, In the proposed system we suggest the permutation algorithm using the output of key steam generator and its security level is analyzed. In this paper we worked out problem of the application about few discrete chaos function through a permutation algorithm and enhanced the security level. Experimental results with image signal demonstrate the proper of the implemented optical encryption system.

• 전기학회논문지
• /
• 제67권1호
• /
• pp.114-123
• /
• 2018

This paper concerns a design and learning method of softmax function neural networks based on K-means clustering. The partial discharge data Information is preliminarily processed through simulation using an Epoxy Mica Coupling sensor and an internal Phase Resolved Partial Discharge Analysis algorithm. The obtained information is processed according to the characteristics of the pattern using a Motor Insulation Monitoring System program. At this time, the processed data are total 4 types that void discharge, corona discharge, surface discharge and slot discharge. The partial discharge data with high dimensional input variables are secondarily processed by principal component analysis method and reduced with keeping the characteristics of pattern as low dimensional input variables. And therefore, the pattern classifier processing speed exhibits improved effects. In addition, in the process of extracting the partial discharge data through the MIMS program, the magnitude of amplitude is divided into the maximum value and the average value, and two pattern characteristics are set and compared and analyzed. In the first half of the proposed partial discharge pattern classifier, the input and hidden layers are classified by using the K-means clustering method and the output of the hidden layer is obtained. In the latter part, the cross entropy error function is used for parameter learning between the hidden layer and the output layer. The final output layer is output as a normalized probability value between 0 and 1 using the softmax function. The advantage of using the softmax function is that it allows access and application of multiple class problems and stochastic interpretation. First of all, there is an advantage that one output value affects the remaining output value and its accompanying learning is accelerated. Also, to solve the overfitting problem, L2-normalization is applied. To prove the superiority of the proposed pattern classifier, we compare and analyze the classification rate with conventional radial basis function neural networks.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2004년도 추계학술대회 논문집 Vol.17
• /
• pp.337-340
• /
• 2004

To develop the highly sensitive Magneto-Impedance sensor, the amorphous ribbon was micro-processed to meander type sensor pattern and the filter circuit was constructed with this pattern. Its external magnetic field dependence of impedance and the output properties of the filter circuit were investigated. The impedance of the pattern had a peak value at the magnetic field of 10 Oe and its changing ratio was about 280%. The impedance change per unit magnetic field was about 36%, in which the output with high sensitivity and linearity could be obtained. The output sensitivity was about 7%/Oe at bias field of 6 Oe..

• 소성∙가공
• /
• 제31권2호
• /
• pp.89-95
• /
• 2022

3D printing products manufactured by material extrusion are used in many industrial fields recently. However, these products are difficult to use in the field due to their low tensile strengths. In order to solve this problem, research on improving the tensile strength of the output using a 3D printer has been continuously conducted. In this study, we performed a tensile test using Universal Testing Machine according to infill pattern, nozzle temperature, bed temperature, and printing speed conditions. Results revealed that tensile specimen of concentric shape had the highest tensile strength in infill pattern condition and that the tensile strength increased linearly with increasing nozzle and bed temperatures. However, the tensile strength decreased with increasing printing speed. Consequently, we confirmed that tensile strength could be increased and decreased depending on output conditions of 3D printing.

• 전기학회논문지
• /
• 제62권5호
• /
• pp.705-711
• /
• 2013

In this paper, we introduce an design of multi-output fuzzy neural networks based on Interval Type-2 fuzzy set. The proposed Interval Type-2 fuzzy set-based fuzzy neural networks with multi-output (IT2FS-based FNNm) comprise the network structure generated by dividing the input space individually. The premise part of the fuzzy rules of the network reflects the individuality of the division space for the entire input space and the consequent part of the fuzzy rules expresses three types of polynomial functions with interval sets such as constant, linear, and modified quadratic inference for pattern recognition. The learning of fuzzy neural networks is realized by adjusting connections of the neurons in the consequent part of the fuzzy rules, and it follows a back-propagation algorithm. In addition, in order to optimize the network, the parameters of the network such as apexes of membership functions, uncertainty factor, learning rate and momentum coefficient were automatically optimized by using real-coded genetic algorithm. The proposed model is evaluated with the use of numerical experimentation.