• 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.

• 여성건강간호학회지
• /
• 제8권4호
• /
• pp.595-607
• /
• 2002

Sexuality education in the period of adolescents need much care and attention. The programs of sexual education through the Internet are excellent resources for adolescents to gain the information related to their sexual health. And systematic program which is necessary for adolescents to manage their sexual health has been rarely found in Korea. The purpose of this study was to offer valuable database for program design and evaluation on sexual education of middle school students through the Internet. Needs assessment for the information of sexual education on the internet among middle school students were carried out. A questionnaire survey was conducted with respondents of 602 middle school students from January to March in 2002. In the sexual counseling center for middle school students, counseling cases through internet were analyzed and evaluated from October, 2001 to September, 2002, We have selected 16 Sexual educational websites in Seoul confirmed Korean Educational Human Resource. Contents which was illustrated in 16 sexual educational websites were analyzed and evaluated by 12 sexual counselors. Design and evaluation of the program on the internet for sexuality education of adolescences was conducted on the basis of this study results by middle school expert teachers, sexual counselors, sexuality education professionals. Data was statistically analyzed using dBSTAT 4.0 for Windows. The extent and phase of the teaching-learning program of the sexuality education on the internet was seen as follows : 1. We evaluated to need for sexuality education on the internet by middle school student. 2. We assessed the properness of sexuality education curriculum on the internet frequently used by middle school students. 3. We designed teaching strategy and learning program for sexuality education of the middle school students. 4. We developed the assessment method for the teaching-learning program of the sexuality education in adolescences on the internet. Middle school students responded that sexual education through Internet is needed in the order of programs related to acquaintances with opposite sex, Sexual culture and ethics, Sexual health, Reproductive health structure and development, Marriage and family, Psychology of Sexuality, Pregnancy and birth. In the internet counseling, cases on the 'reproductive health structure and development' was ranked as the top. In short we have found the most needs as follows; Meaning of the marriage life and having family, Sexes and Love, Human relation, Sexual Culture.We recommend as follows on the basis of this study results: 1. It is necessary for sexuality education program on the internet to specify according to age and target the specific individual needs. 2. Sexual educators have to employ various educational materials such as flash, cartoon, multimedia in order to provide effective sexuality education. 3. Internet based sex education need to be evaluated regularly through reassessment of the effectiveness of sexuality education for content quality and richness.

• 한국과학교육학회지
• /
• 제30권5호
• /
• pp.533-543
• /
• 2010

이 연구는 자유주제 과학탐구 과정에서 일어나는 인지수준 향상 정도 및 탐구단계별 상호작용을 분석함으로써, 자유주제 과학탐구에 대한 질적 이해를 돕고자 하였다. 자유주제 과학탐구는 인지수준이 과도기(2B/3A)이고, 학습동기수준이 심층학습형, 심층능력형인 학생들의 탐구능력 신장에 긍정적인 효과가 있었다. 또한 학생들은 탐구과정 중 주제 및 가설설정단계에서 어려움을 겪으며, 자유탐구가 진행됨에 따라 논의의 기술은 증가하나, 상호작용의 질적 향상은 쉽게 나타나지 않음을 알 수 있었다. 심층학습형은 모둠 내에서 사고의 정교화를 촉진하는 역할을 하고 있었다. 심층능력형은 가장 주도적으로 상호작용에 참여하여 논의를 이끄나, 탐구결과에 대한 집착으로 과정적 측면을 소홀히 하는 경향이 있었다. 그리고 피상능력형은 하위수준의 상호작용이 많아서 논의를 심화시키지 못하나, 자유탐구가 진행됨에 따라 점차 상위수준의 상호작용이 증가하는 긍정적인 변화를 보였다. 이 연구로부터 자유주제 과학탐구 활동의 특성 및 자유주제 과학탐구 활동의 진행과 모둠구성에 대한 시사점을 얻을 수 있었으며, 이에 대해 논하였다.

• 수면정신생리
• /
• 제9권1호
• /
• pp.41-47
• /
• 2002

목 적 : 여러 모의 연구(simulated study)에서 순환제 교대근무자의 야간 근무에 따른 수면장애와 작업수행의 어려움에 대해 광치료가 치료적 도구로 제안되고 있어 실제 순환제 교대 근무 환경에서의 야간 근무 동안 광치료의 효과를 평가하고자 하였다. 방 법 : 연구대상인 5명의 병동 근무 간호사들은 기초평가단계 동안 연속적인 3일간의 낮 근무 후 650 lux의 실내 조명 하에서 연속적인 3일간의 야간근무를 하도록 계획되었으며, 2주 후의 광치료시행단계에서는 동일한 순환주기로 근무하는 동안 자정부터 오전 4시까지 2500 lux의 광박스를 이용한 광치료를 시행하였다. 야간근무 후의 낮수면은 actiwatch를 이용하여 평가되었으며 야간 근무 동안은 스탠포드 졸리움 척도를 이용한 각성상태, 고막체온, STIM을 이용한 수행능력 및 visual analogue를 이용한 우울감, 불안감, 신체적 불편감, 피로감을 기초평가단계와 광치료시행단계 각각에서 야간근무 첫째날과 셋째날을 비교평가하였다. 결 과 : 야간근무 후의 낮 수면은 기초평가단계와 광치료 시행단계 모두 야간 근무 1일째와 3일째에 유의한 차이를 보이지 않았다. 일주기리듬의 위상 변화를 알아보기 위해 측정된 고막 온도는 야간근무 3일째에 기초평가단계에서는 위상전진을 나타낸 반면 광치료시행단계에서는 위상변화가 없었다. 또한, 각성상태, 우울감, 불안감, 피로감, 신체적 불편감, 졸리움 등의 변화도 기초평가단계에서 야간근무 3일째에 유의하게 악화된 반면 광치료시행단계에서는 변화가 없었다. 주의력에서는 광치료시행단계에서만 야간근무 3일째가 1일째에 비해 유의하게 호전되는 양상을 나타내었다. 결 론 : 본 연구는 위약효과를 배제할 수는 없지만, 실제근무 환경에서 단기간의 광치료가 순환제 교대근무자들의 주간 수면의 질을 뚜렷하게 변화시키진 않았으나, 야간 근무동안 졸리움 감소와 주의력 향상을 통한 야간 업무 수행능력의 호전 가능성을 제시해 주고 있다.

• International Journal of Quality Innovation
• /
• 제6권1호
• /
• pp.109-118
• /
• 2005

The parameter design is the most emphasized measure by researchers for a new products development. It is critical for makers to achieve simultaneously in both the time-to-market production and the quality enhancement. However, there are difficulties in practical application, such as (1) complexity and nonlinear relationships co-existed among the system's inputs, outputs and control parameters, (2) interactions occurred among parameters, (3) where the adjustment factors of Taguchi's two-phase optimization procedure cannot be sure to exist in practice, and (4) for some reasons, the data became lost or were never available. For these incomplete data, the Taguchi methods cannot treat them well. Neural networks have a learning capability of fault tolerance and model free characteristics. These characteristics support the neural networks as a competitive tool in processing multivariable input-output implementation. The successful fields include diagnostics, robotics, scheduling, decision-making, prediction, etc. This research is a case study of spherical annealing model. In the beginning, an original model is used to pre-fix a model of parameter design. Then neural networks are introduced to achieve another model. Study results showed both of them could perform the highest spherical level of quality.

• 동력기계공학회지
• /
• 제7권2호
• /
• pp.29-35
• /
• 2003

The monitoring and diagnostics of the rotating machinery have been received considerable attention for many years. The objectives are to classify the machinery condition and to find out the cause of abnormal condition. This paper describes a classification method of diagnosing the small reciprocating compressor for refrigerators using the artificial neural network and the wavelet transform. In order to extract salient features, the wavelet transform are used from primary noise signals. Since the wavelet transform decomposes raw time-waveform signals into two respective parts in the time space and frequency domain, more and better features can be obtained easier than time-waveform analysis. In the training phase for classification, self-organizing feature map(SOFM) and learning vector quantization(LVQ) are applied, and the accuracies of them ate compared with each other. This paper is focused on the development of an advanced signal classifier to automatize the vibration signal pattern recognition. This method is verified by small reciprocating compressors, for refrigerator and normal and abnormal conditions are classified with high flexibility and reliability.

• 반도체디스플레이기술학회지
• /
• 제14권2호
• /
• pp.29-34
• /
• 2015

Traffic lights are common in cities and are important cues for the path planning of intelligent vehicles. In this paper, we propose a robust and efficient algorithm for recognizing traffic lights from video sequences captured by a low cost off-the-shelf camera. Instead of using color information for recognizing traffic lights, a shape based approach is adopted. In learning and detection phase, Histogram of Oriented Gradients (HOG) feature is used and a cascade classifier based on Adaboost algorithm is adopted as the main classifier for locating traffic lights. To decide the color of the traffic light, a technique based on histogram analysis in HSV color space is utilized. Experimental results on several video sequences from typical urban environment prove the effectiveness of the proposed algorithm.

• 예술인문사회 융합 멀티미디어 논문지
• /
• 제5권2호
• /
• pp.183-190
• /
• 2015

본 연구에서는 기억 책략을 사용하여도 이에 따른 이득을 볼 수 없는 현상, 즉 책략 사용이 증가하는 만큼 회상량이 증가하지 않는 이용결여 현상을 훈련 상황이 아닌 자연스런 상황에서 알아보고자 하였다. 초등학교 5학년 학생을 대상으로 5회기 동안 단어쌍 기억 과제에서의 정교화 책략을 연구하였다. 연구 결과, 회기가 진행되면서 정교화 책략 사용량에는 증가를 보였지만 회상량은 회기에 따른 유의미한 증가가 나타나지 않아 이용결여 현상이 나타났다. 이용결여 현상은 아동이 효율적인 책략 사용의 단계로 발달해 가는데 필요한 적응적 과정임을 시사한다.

• 한국전산구조공학회논문집
• /
• 제37권2호
• /
• pp.143-149
• /
• 2024

이미지 분석을 통한 재료의 상 구분은 재료의 미세구조 분석을 위해 필수적이다. 이미지 분석에 주로 사용되는 마이크로-CT 이미지는 대체로 재료를 구성하고 있는 상에 따라 회색조 값이 다르게 나타나므로 이미지의 회색조 값 비교를 통해 상을 구분한다. 순환골재의 고체상은 수화된 시멘트풀과 천연골재로 구분되는데, 시멘트풀과 천연골재는 CT이미지 상에서 유사한 회색조 분포를 보여 상을 구분하기 어렵다. 본 연구에서는 Unet-VGG16 네트워크를 활용하여 순환골재 CT 이미지로부터 천연골재를 분할하는 자동화 방법을 제안하였다. 딥러닝 네트워크를 활용하여 2차원 순환골재 CT 이미지로부터 천연골재 영역을 분할하는 방법과 이를 3차원으로 적층하여 3차원 천연골재 이미지를 얻는 방법을 제시하였다. 선별된 3차원 천연골재 이미지에서 각각의 골재 입자를 분할하기 위해 이미지 필터링을 사용하였다. 골재 영역 분할 성능을 정확도, 정밀도, 재현율 F1 스코어를 통해 검증하였다.

• Journal of Chest Surgery
• /
• 제25권11호
• /
• pp.1362-1368
• /
• 1992

Between March 1986 and November 1992, thirty-one patients with left main coronary artery stenosis[LMCAS, over 50% of cross sectional area] were revascularized at Seoul National University Hospital. The incidence of LMCAS was 20.8%[31/149]. The male:female ratio was 15:16. Age ranged from 39 to 70 years, with a mean age of 51 years. The anginal syndrome was composed of 23 unstable, 6 stable and 2 post-infarction angina preoperatively. There were sixteen isolated LMCAS, four ostial stenosis and eleven combined distal and /or right coronary artery stenosis. The degree of LMCAS was 50-74% in 21 patients[67.7%], 75-89% in one[3.6%] and 90-99% in 9[29.1%]. There was no case with 100% obstruction. Of the total patients with LMCAS, 11 patients received 4 distal anastomoses, another 11 patients had 3 distal anastomoses, and 8 patients needed 2 distal anastomoses. The overall operative mortality was 12.9%[4 /31], and the incidence of which was higher than the remaing group [6.8%, 8/118]. The causes of death were myocardial infarction[2 patients], ventricular arrhythmia[1 patient] and brain damage[1 patient]. All patients have been followed-up for average 28.9 months[1-76 months]. There was no late death. But one patients experinced anginal recurrence. In conclusion, making allowance for its notorious clinical results and relatively higher incidence in Korea, aggressive surgical techniques such as retrograce myocardial perfusion may be mandatory while we are in a learning phase.