• Plant Journal
• /
• v.12 no.2
• /
• pp.36-40
• /
• 2016

At the contract for power plant construction, the penalty appropriation on performance decrease is signed between ordering organization and construction firm. In this, the penalty cost signed must be reasonable value that both of ordering organization and construction firm can accept, therefore the methodology for penalty appropriation is very important. Cogeneration is a system that produces electricity and heat at the same time, therefore the penalty appropriation for cogeneration should be uncertain. Thermoeconomics analyzes various energy costs, however the relation of thermoeconomics and penalty cost may not be analyzed up to now. The aim of this study demonstrates that thermoeconomics can be applied to the penalty appropriation at the performance acceptance test. As the result of CGAM system, if the construction cost is $10,000,000, the value of $6,665,688 was appropriated to the electricity production performance and the value of $3,334,312 was appropriated to the heat production performance. Therefore if one percentage at the electricity production performance decreases, the penalty is $6,666, and one percentage at the heat production performance decrease, we can understand that the penalty is $3,334.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.4
• /
• pp.721-731
• /
• 2017

In this study, principal component analysis methods using Lasso penalty are introduced. There are two popular methods that apply Lasso penalty to principal component analysis. The first method is to find an optimal vector of linear combination as the regression coefficient vector of regressing for each principal component on the original data matrix with Lasso penalty (elastic net penalty in general). The second method is to find an optimal vector of linear combination by minimizing the residual matrix obtained from approximating the original matrix by the singular value decomposition with Lasso penalty. In this study, we have reviewed two methods of principal components using Lasso penalty in detail, and shown that these methods have an advantage especially in applying to data sets that have more variables than cases. Also, these methods are compared in an application to a real data set using R program. More specifically, these methods are applied to the crime data in Ahamad (1967), which has more variables than cases.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.6
• /
• pp.941-947
• /
• 2018

Station transfers for smart card based railway networks refer to transfer pedestrian movements that occur at the origin and destination nodes rather than at a middle station. To calculate the optimum path for the railway network, a penalty for transfer pedestrian movement must be included in addition to the cost of within-car transit time. However, the existing link label-based path searching method is constructed so that the station transfer penalty between two links is detected. As such, station transfer penalties that appear at the origin and destination stations are not adequately reflected, limiting the effectiveness of the model. A ghost node may be introduced to expand the network, to make up for the station transfer penalty, but has a pitfall in that the link label-based path algorithm will not hold up effectively. This research proposes an optimal path search algorithm to reflect station transfer penalties without resorting to enlargement of the existing network. To achieve this, a method for applying a directline transfer penalty by comparing Ticket Gate ID and the line of the link is proposed.

• Plant Journal
• /
• v.12 no.4
• /
• pp.32-36
• /
• 2016

In this study, a penalty cost appropriation methodology for CGAM performance acceptance test was suggested. As the result of CGAM performance test, there were 0.31% decreases in electricity output, 0.39% decreases in heat rate on electricity output, 0.31% increases in heat output, and 0.23% increases in heat rate on heat output. As the result of penalty cost appropriation for above performance, the penalty cost was calculated as -$20,837 in electricity output, -$25,930 in heat rate on electricity output, +$10,340 in heat output, and +$7,715 in heat rate on heat output. Each penalty is appropriated as above fore kinds, however the total penalty should be determined as how to combine above fore kinds of penalty. In our calculation, the minimum total penalty was -$18,215 and the maximum total penalty was -$46,767. The methodology of total penalty appropriation should be determined in the contract between ordering organization and construction firm, and we can understand that it is very important.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.11a
• /
• pp.209-212
• /
• 2009

본 논문에서는 MDCT기반의 오디오 특징과 영상 특징을 이용하여 축구 동영상의 하이라이트를 효과적으로 요약하는 방식을 제안한다. 제안하는 방식에서는 입력되는 축구 동영상을 비디오 신호와 오디오 신호로 분리한 후에, 분리된 연속적인 오디오 신호를 압축영역의 MDCT계수를 통해 이벤트 사운드별로 분류하여 오디오 이벤트 후보구간을 추출한다. 입력된 비디오 신호에서는 장면 전환점을 추출하고 추출된 장면 전환점으로부터 페널티 영역을 검출한다. 검출된 오디오 이벤트 후보구간과 검출된 페널티 영역장면을 함께 결합하여 축구 동영상의 이벤트 장면을 검출한다. 검출된 페널티 영역 장면을 통해 검출된 이벤트 구간을 다른 이벤트 구간보다 더 높은 우선순위를 갖는 하이라이트로 선정하여 요약본이 생성된다. 생성된 하이라이트 요약본의 평가는 precision과 recall을 통해 정확도를 평가하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04b
• /
• pp.253-255
• /
• 2002

본 논문에서는 한글, 숫자, 영문자, 기호 통이 혼용된 컬러용 품지 또는 A4지 처방전을 자동으로 인식하는 시스템을 설계하는 방법을 제안한다. 이를 구현하기 위해 먼저 처방전을 스캐너를 이용하여 스캔하고 컬러 정보를 이용하여 회전된 처방전을 보정한 다음 처방전의 종류를 결정한다. 그리고, 문자의 형태학적 특징에 따라 한글과 그 외의 문자(비한글)를 구분한다. 그리고, 구분된 비한글의 경우는 다양할 특징 벡터를 이용한 가중 페널티 방법을 이용하여 인식한다

• The KIPS Transactions:PartB
• /
• v.8B no.2
• /
• pp.173-182
• /
• 2001

본 논문에서는 투과형 토머그래피 영상재구성을 위하여 EM 알고리즘을 사용하는 경우에 발생하는 문제점을 해결할 수 있는 방안을 제시한다. 일반적으로 토머그래피 영상재구성과 같은 다-차원의 모수 추정인 경우에서는 그것의 페널티 함수의 헤이지안행렬의 역행렬 차수가 매우 높기 때문에 그것을 직접적으로 계산할 수 없다. 이러한 문제점을 해결하기 위하여 PEMG-1 알고리즘을 제안한다. 이 알고리즘은 페널티 함수를 사용하는 그래디언트 형태의 알고리즘인데 이것은 Lange(1995)과 Green(1990)의 알고리즘에서 지적된 문제점을 동시에 해결할 수 있다.

• Korean Journal of Applied Biomechanics
• /
• v.19 no.3
• /
• pp.499-508
• /
• 2009

The purpose of this study was to compare and analyze flick shooting motion in penalty corner between high school players and national players in field hockey. Five high school players and six national players participated in this study. The 3D kinematic data were collected for each subject performing the penalty corner stroke. The results of the study were as follows: 1) The national players had higher stick head and ball velocity than the high school players. 2) The forward length between ball and support foot during ball catching with stick head was longer in the national players than the high school players. 3) At the Z axis of the E5 event, the center of gravity of the national players was lower than that of the high school players. 4) At the Z axis of the E5 event, left hip angle of the national players was lower than that of the high school players. 5) The national players had longer drag length of ball than the high school players. 6) The national players had higher hand and lower arm angular momentum than the high school players.

• Journal of the Korea Society for Simulation
• /
• v.19 no.4
• /
• pp.1-9
• /
• 2010

In this paper, advanced evacuation analysis simulation on a passenger ship is performed. Velocity based model has been implemented and used to calculate the movement of the individual passengers under the evacuation situation. The age and gender of each passenger are considered as the factors of walking speed. Flocking algorithm is applied for the passenger's group behavior. Penalty walking velocity is introduced to avoid collision between the passengers and obstacles, and to prevent the position overlap among passengers. Application of flocking algorithm and penalty walking velocity to evacuation simulation is verified through implementation of the 11 test problems in IMO (International Maritime Organization) MSC (Maritime Safety Committee) Circulation 1238.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.301-306
• /
• 2019

문서 생성 요약은 최근 딥러닝을 이용한 end-to-end 시스템을 통해 유망한 결과들을 보여주고 있어 연구가 활발히 진행되고 있는 자연어 처리 분야 중 하나이다. 하지만 문서 생성 요약 모델을 구성하기 위해서는 대량의 본문과 요약문 쌍의 데이터 셋이 필요한데, 이를 구축하기가 쉽지 않다. 따라서 본 논문에서는 정교한 뉴스 기사 요약 데이터 셋을 기계적으로 구축하는 방법을 제안한다. 또한 딥러닝 기반의 생성 요약은 입력 문서와 다른 정보를 생성하거나, 또는 같은 단어를 반복하여 생성하는 문제점들이 존재한다. 이를 해결하기 위해 요약문을 생성할 때 입력 문서의 내용을 인용하는 복사-메커니즘과, 추론 단계에서 단어 반복을 직접적으로 제어하는 페널티를 사용하면 상대적으로 안정적인 문장이 생성될 수 있다. 그리고 Transformer 모델은 순환 신경망 모델보다 요약문 생성 과정에서 시퀀스 길이가 긴 본문의 정보를 적절히 인코딩하여 줄 수 있는 모델이다. 따라서 본 논문에서는 복사-메커니즘과 추론 단계의 페널티를 이용한 Copy-Transformer 모델을 한국어 문서 생성 요약 데이터에 적용하였다. 네이버 지식iN 질문 요약 데이터 셋과 뉴스 기사 요약 데이터 셋 상에서 실험한 결과, 제안한 모델을 이용한 생성 요약이 비교 모델들 대비 가장 좋은 성능을 보이고 양질의 요약을 생성하는 것을 확인하였다.