• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.5
• /
• pp.594-600
• /
• 2006

The optimization has been performed to search an economical running pattern in the view point of trip time and energy consumption. Fuzzy control model has been applied to build the meta-model. To identify the structure and its parameters of a fuzzy model, fuzzy c-means clustering method and differential evolutionary scheme ate utilized, respectively. As a result, two meta-models for trip time and energy consumption are constructed. The optimization to search an economical running pattern is achieved by differential evolutionary scheme. The result shows that the proposed methodology is very efficient and conveniently applicable to the operation of railway system.

• Annual Conference of KIPS
• /
• 2012.11a
• /
• pp.1357-1360
• /
• 2012

온라인 쇼핑객의 구매 규칙을 예견하기 위해 기업은 데이터 마이닝 기법을 사용하는데, 최대 빈발 패턴은 특정한 고객의 구매 원칙을 드러내기 때문에, 최대 빈발 패턴에 대한 마이닝은 최근 시장 분석에서 핵심적 이슈가 되고 있다. 본 논문에서 우리는 오리지널 데이터세트로부터 널 트랜잭션(Null Transaction)을 제거한 후, 최대 빈발 패턴을 발생시키기 위한 BRE-트리(Bottom-up Row Enumeration Tree)를 적용시켰다. 다음으로 온라인 거래 데이터베이스에서 고객 구매 규칙의 마이닝을 위한 항목들 간의 거리를 계산하기 위해, SCL(Sequence Close Level)의 변형된 버전을 사용하였다. 실험결과는 합리적인 시간 내에 고객의 구매 규칙을 더 정확하게 예견할 수 있음을 보여준다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.3
• /
• pp.125-137
• /
• 2009

Acoustic emission-based techniques are being used for the nondestructive inspection of mechanical systems used in service. For reliable fault monitoring related to the crack growth, it is important to identify the dynamical characteristics as well as transient crack-related signals. Widely used methods which are based on physical phenomena of the three damage stages for detecting the crack growth have a problem that crack-related acoustic emission activities overlap in time, therefore it is insufficient to estimate the exact crack growth time. The proposed pattern recognition method uses the dynamical characteristics of acoustic emission as inputs for minimizing false alarms and miss alarms and performs the temporal clustering to estimate the crack growth time accurately. Experimental results show that the proposed method is effective for practical use because of its robustness to changes of acoustic emission caused by changes of pressure levels.

• Journal of The Korean Association of Information Education
• /
• v.17 no.3
• /
• pp.347-355
• /
• 2013

A robot assisted education has advantages to increase the students' flow degree. Especially, when we teach programming with robots we take advantages that the students can easily understand the programs because they are interested in the robots and they can see the movement of the robots. However, it is difficult to teach programming to meet our purposes because the students have reluctances to the robot programming with sensors and they take a lot of time building the robots. Therefore, in this paper I analysed the patterns of the robot programming and propose new robot programming education methods that the students need not to care to the sensors. According to this methods, we can teach the robot programming efficiently because we reduce the time to build and programming the robots.

• Journal of KIISE:Computing Practices and Letters
• /
• v.12 no.5
• /
• pp.312-325
• /
• 2006

Integration applications, such as EAI, B2Bi, need stable massive data processing systems during overload state cause by service request congestion in a short period time. In this paper, we propose the PLC (Peak Load Control)-based Worker-Linker pattern, which can effectively and stably process massive I/O transactions in spite of overload state generated by service request congestion. This pattern uses the delay time algorithm for the PLC mechanism. In this paper, we also show the example of applying the pattern to business-business integration framework and the experimental result for proving the stability of performance. According to our experiment result, the proposed delay time algorithm can stably control the heavy overload after the saturation point and has an effect on the controlling peak load.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.2
• /
• pp.395-406
• /
• 2017

The purpose of this study is to identify the pattern of daily electricity demand through clustering and classification. The hourly data was collected by KPS (Korea Power Exchange) between 2008 and 2012. The time trend was eliminated for conducting the pattern of daily electricity demand because electricity demand data is times series data. We have considered k-means clustering, Gaussian mixture model clustering, and functional clustering in order to find the optimal clustering method. The classification analysis was conducted to understand the relationship between external factors, day of the week, holiday, and weather. Data was divided into training data and test data. Training data consisted of external factors and clustered number between 2008 and 2011. Test data was daily data of external factors in 2012. Decision tree, random forest, Support vector machine, and Naive Bayes were used. As a result, Gaussian model based clustering and random forest showed the best prediction performance when the number of cluster was 8.

• Journal of Energy Engineering
• /
• v.5 no.2
• /
• pp.183-192
• /
• 1996

This paper presents a numerical model of the optimal operation pattern of the CHP system built for duel-purpose of power generation and district heat production. The model can be differently formulated in accordance with the view of planner: society, electric utility or district-heating company. Here, the operation pattern of the system components and the effect of heat price are of major interest in the study. From the case study, it was found that the optimal use of auxiliary heating equipment is very important to achieve the minimum societal cost. And, the multi-step heat pricing scheme is desirable to induce the voluntary behavior of both companies towards the societal optimal pattern.

• Journal of Korea Multimedia Society
• /
• v.5 no.3
• /
• pp.263-271
• /
• 2002

In this paper, we propose a new smoothing method for removing moire patterns using analysis of halftone patterns interference. The proposed method can determine a strength of moire patterns by using gray values of pixels and the size of smoothing mask for moire region is adjusted adaptively according to the strength of moire patterns. Therefore it can remove moire patterns effectively and preserve meaningful high frequencies well, such as edges and textures. The proposed method only refer to predefined lookup table to determine the strength of moire patterns, so it is more efficient than a previous work based on FFT of subblock. It could be applied to field of various multimedia applications that deal with color prints.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2001.11a
• /
• pp.158-163
• /
• 2001

본 논문의 국내 패션 업체에서 상품기획 과정 중 많은 시간과 노력을 투자해야 했던 디자인 및 패턴 자료를 데이터베이스화하여 key word를 통해 효율적으로 찾아 사용할 수 있도록 한 디지털 여성 자켓 분류 데이터베이스 시스템을 개발하고자 실시되었다. 이를 위해 의류업체 종사자 48명과 의류학 전공 대학원생 54명, 패턴 전문가 11명을 대상으로 설문조사 및 인터뷰를 실시하여 디자인 및 패턴의 분류 기준과 의류 생산기획 업무의 디지털화 가능성을 검토하였다. 본 연구의 결과는 다음과 같다. 국내 의류업체에서는 상품기획시 국외패션잡지와 collection지를 가장 많이 활용하고 있었으며, 디자인 및 패턴 DB 프로그램에 대한 효용성 기대와 수용도 기대에는 집단간 유의한 차이가 있는 것으로 나타났다. 여성 자켓 디자인 DB를 위한 구성요소 분류 기준은 7가지고 선정되었고, 이미지 형용사 분류 기준은 6가지로 선정되었다. 또한 자켓 제작을 위한 block pattern 분류 기준은 4가지로 선정되었다. 본 연구를 통해 개발된 자켓 디자인 선택 프로그램의 모델을 제시한 후 실험 참가자들에게 효용성 및 사용가능성을 다시 검증한 결과, 프로그램 제시 전 조사결과보다 유의하게 긍정적으로 평가되었으며, 데이터베이스 자료 활용시 이미지 형용사를 통한 검색보다는 구성요소를 통한 검색에 더 만족하는 것으로 나타났다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2008.05a
• /
• pp.2061-2065
• /
• 2008

본 연구는 강우-유출 및 TOC의 패턴 분류를 위하여 광주 광산 강우관측소의 강우량자료와 나주지점의 유출량 그리고 기존의 BOD 및 COD 수질농도 측정값에 비하여 적은 오차요인과 빠른 시간에 결과 값을 얻을 수 있으며 유출량과 난분해성 물질에 대한 해석이 가능하고 재현성이 탁월한 TOC자료를 사용하였다. SOM을 적용하기 위해 먼저 Map의 크기는 Garcia가 제시한 $M=5{\sqrt{N}}$을 이용하여 결정한다. 이러한 비선형적인 다변량 자료를 분석하기 위해서 Map에 의해 구분된 자료 위치를 추출하여 원자료를 재구축하고 이를 통해 원자료를 패턴별로 분류 할 수 있었다. 이러한 패턴별 분류를 통해 유출량에 따른 TOC자료를 2차원의 Map 상에 시각적으로 가시화하여 비선형적인 경향이 강한자료의 분포적 양상을 이해하는데 큰 도움이 되며, 향후 이를 통해 예측을 위한 모형화 과정에도 크게 도움을 줄 것으로 기대된다. 또한, 강우자료 또는 유출량 자료만을 이용한 단일변량의 패턴분류를 위해 SOM의 적용이 가능할 것으로 판단되며, 이는 각 변량의 본질적인 특성을 파악할 수 있을 것으로 기대된다.