• 정보과학회 논문지
• /
• 제42권9호
• /
• pp.1124-1132
• /
• 2015

일반적으로, 교사 학습 알고리즘이 적절히 학습되기 위해서는 레이블의 편향이 없는 충분한 양의 학습데이터가 필요하다. 그러나 영작문 자동채점 시스템 개발을 위한 충분하고 편향되지 않은 학습데이터를 수집하는 것은 어려운 일이다. 또한 영어 작문 평가의 경우, 전체적인 답안 수준에 대한 다면적인 평가가 이루어진다. 적고 편향되기 쉬운 학습데이터와 이를 이용한 여러 평가영역에 대한 학습모델을 생성해야하기 때문에, 이를 위한 적절한 기계학습 알고리즘을 결정하기 어렵다. 본 논문에서는 이러한 문제를 앙상블학습을 통해 완화할 수 있음을 실험에 통해 보이고자 한다. 실제 중, 고등학교 학생들을 대상으로 시행된 단문형 영작문 채점 결과를 학습데이터 개수와 편향성을 조절하여 실험하였다. 학습데이터의 개수 변화와 편향성 변화의 실험 결과, 에이다부스트 알고리즘을 적용한 결과를 투표로 결합한 앙상블 기법이 다른 알고리즘들 보다 전반적으로 더 나은 성능을 나타냄을 실험을 통해 나타내었다.

• 한국해양공학회지
• /
• 제33권5호
• /
• pp.427-434
• /
• 2019

Artificial intelligence (AI)-aided research currently enjoys active use in a wide array of fields thanks to the rapid development of computing capability and the use of Big Data. Until now, forecasting methods were primarily based on physics models and statistical studies. Today, AI is utilized in disaster prevention forecasts by studying the relationships between physical factors and their characteristics. Current studies also involve combining AI and physics models to supplement the strengths and weaknesses of each aspect. However, prior to these studies, an optimization algorithm for the AI model should be developed and its applicability should be studied. This study aimed to improve the forecast performance by constructing a model for neural network optimization. An artificial neural network (ANN) followed the ever-changing path of a typhoon to produce similar typhoon predictions, while the optimization achieved by the neural network algorithm was examined by evaluating the activation function, hidden layer composition, and dropouts. A learning and test dataset was constructed from the available digital data of one typhoon that affected Korea throughout the record period (1951-2018). As a result of neural network optimization, assessments showed a higher degree of forecast accuracy.

• 시스템엔지니어링학술지
• /
• 제14권2호
• /
• pp.24-32
• /
• 2018

System Complexity one of the most common cause failure of the projects, it leads to a lack of understanding about the functions of the system. Hence, the model is developed for communication and furthermore modeling help analysis, design, and understanding of the system. On the other hand, the text-based specification is useful and easy to develop but is difficult to visualize the physical composition, structure, and behaviour or data exchange of the system. Therefore, it is necessary to transform system description into a diagram which clearly depicts the behaviour of the system as well as the interaction between components. According to the International Atomic Energy Agency (IAEA) Safety Glossary, The safety system is a system important to safety, provided to ensure the safe shutdown of the reactor or the residual heat removal from the reactor core, or to limit the consequences of anticipated operational occurrences and design basis accidents. Core Protection Calculator System (CPCS) in Advanced Power Reactor 1400 (APR 1400) Nuclear Power Plant is a safety critical system. CPCS was developed using systems engineering method focusing on Departure from Nuclear Boiling Ratio (DNBR) calculation. Due to the complexity of the system, many diagrams are needed to minimize the risk of ambiguities and lack of understanding. Using Model-Based Systems Engineering (MBSE) software for modeling the DNBR algorithm were used. These diagrams then serve as the baseline of the reverse engineering process and speeding up the development process. In addition, the use of MBSE ensures that any additional information obtained from auxiliary sources can then be input into the system model, ensuring data consistency.

• 한국차세대컴퓨팅학회논문지
• /
• 제14권6호
• /
• pp.15-29
• /
• 2018

본 논문에서는 기존 파티클 스웜 최적화를 기반으로 추적 대상 객체의 이동 궤적을 이용하는 객체 추적기에서 시간 정보 활용의 문제점을 개선한 강인한 객체 추적 알고리즘을 제안한다. 제안하는 알고리즘은 추적 대상 객체와 유사한 특징을 가지는 변위들의 집합에 대한 위치들의 온라인 업데이트와 추적을 가능하게 한다. 객체들의 중첩을 검출하고 추적 대상의 위치를 결정하기 위해 궤적 정보와 변위들의 집합을 기반으로 적응적 파라미터를 사용하는 규칙기반 접근을 사용한다. 기존 알고리즘들과 비교해보면 제안하는 접근법은 가용한 정보를 복합적으로 사용함으로써 각종 임계값에 대한 적응적 조정을 가능하게 한다. 또한, 파티클 스웜 최적화에서 발산에 의한 손실과 불완전한 수렴의 문제를 해결하기 위해 효율적인 가중치 조절 함수를 제안하고 있다. 제안하는 가중치 조절 함수는 파티클들이 최적의 해에 수렴하기 이전에 전체 프레임 영역에서 탐색할 수 있도록 한다. 유사한 특징 조합을 가지는 다중 객체가 존재하는 환경에서 제안 알고리즘을 테스트한 결과, 기존 스웜 최적화 기반의 객체 추적기들에 비해 기존 유사 변위들에 대한 잘못된 추적을 현저히 줄이는 것을 확인할 수 있었다.

• 정보과학회 논문지
• /
• 제44권9호
• /
• pp.908-917
• /
• 2017

웹 서비스 기술이 각광받고 그 사용이 확대됨에 따라, 복잡하고 동적인 서비스 환경에서 사용자에게 적절한 서비스를 추천하는 방법에 대한 연구가 활발히 진행되고 있다. 또한 효과적인 서비스 매쉬업 개발을 위해 서비스를 추천하는 방법이 제안되었으나, 기존의 매쉬업 단위 서비스 추천 방식은 여러 매쉬업 개발자의 성향을 분석하여 그에 맞는 서비스를 추천하지는 못하였다. 이에 본 논문에서는 매쉬업 개발자들이 만든 서비스 매쉬업의 집합들과 추천 대상 개발자의 매쉬업 집합 사이의 유사도를 측정하고 유사한 매쉬업 집합들로부터 서비스를 추천하는 방법을 제안한다. 그리고 ProgrammableWeb에서 수집된 매쉬업 데이터로 실험한 결과를 비교 분석하여 본 연구의 방법이 사용자 기반 협업 필터링 알고리즘보다 높은 정확도와 재현율을 보임을 확인하였다.

• 전력전자학회논문지
• /
• 제21권5호
• /
• pp.453-456
• /
• 2016

The major causes of electrical fire in low-voltage distribution lines are classified into short-circuit fault, overload fault, electric leakage, and electric contact failure. The special principal factor of the fire is electric arc or spark accompanied with such electric faults. This paper studies the development of an electric fire prevention system with detection and alarm of that in case of parallel arc fault occurrence in low-voltage distribution lines. The proposed system is designed on algorithm sensing the instantaneous voltage drop of line voltage at arc fault occurrence. The proposed detector has characteristics of high-speed operation responsibility and superior system reliability from composition using a large number of semiconductor devices. A new sensing control method that shows the detection of parallel arc fault is sensed to ripple voltage drop through a diode bridge full-wave rectifier at electrical accident occurrence. Some experimental tests of the proposed system also confirm the practicality and validity of the analytical results.

• 대한전기학회논문지:전기기기및에너지변환시스템부문B
• /
• 제53권3호
• /
• pp.152-163
• /
• 2004

This paper presents a sensorless speed control of cylindrical permanent magnet synchronous motors(PMSM) using an adaptive integral binary observer In view of composition with a main loop regulator and an auxiliary loop regulator, the binary observer has a property of the chattering alleviation in the constant boundary layer. However, the steady state estimation accuracy and robustness are dependent upon the width of the constant boundary. In order to improve the steady state performance of the binary observer, the binary observer is formed by adding extra integral dynamics to the switching hyperplane equation. With the help of integral characteristic, the rotor speed can be finely estimated and utilized for a sensorless speed controller for PMSM. Since the Parameters of the dynamic equations such as machine inertia or a viscosity friction coefficient are lot well known, there are many restrictions in the actual implementation. The proposed adaptive integral binary observer applies an adaptive scheme so that observer may overcome the problem caused by using the dynamic equations and the rotor speed is constructed by using the Lyapunov function. The observer structure and its design method are described. The experimental results of the proposed algorithm are presented to demonstrate the effectiveness of the approach.

• 대한기계학회논문집A
• /
• 제26권6호
• /
• pp.1092-1098
• /
• 2002

This paper discusses the composition of the theory of reinforcement teaming, which is applied in real-time teaming, and evolutionary strategy, which proves its the superiority in the finding of the optimal solution at the off-line teaming method. The individuals are reduced in order to team the evolutionary strategy in real-time, and new method that guarantee the convergence of evolutionary mutations are proposed. It is possible to control the control object varied as time changes. As the state value of the control object is generated, applied evolutionary strategy each sampling time because of the teaming process of an estimation, selection, mutation in real-time. These algorithms can be applied, the people who do not have knowledge about the technical tuning of dynamic systems could design the controller or problems in which the characteristics of the system dynamics are slightly varied as time changes. In the future, studies are needed on the proof of the theory through experiments and the characteristic considerations of the robustness against the outside disturbances.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권4호
• /
• pp.1579-1602
• /
• 2020

In the design of production system, buffer capacity allocation is a major step. Through polymorphism analysis of production capacity and production capability, this paper investigates a buffer allocation optimization problem aiming at the multi-stage production line including unreliable machines, which is concerned with maximizing the system theoretical production rate and minimizing the system state entropy for a certain amount of buffers simultaneously. Stochastic process analysis is employed to establish Markov models for repairable modular machines. Considering the complex structure, an improved vector UGF (Universal Generating Function) technique and composition operators are introduced to construct the system model. Then the measures to assess the system's multi-state reliability and structural complexity are given. Based on system theoretical production rate and system state entropy, mathematical model for buffer capacity optimization is built and optimized by a specific genetic algorithm. The feasibility and effectiveness of the proposed method is verified by an application of an engine head production line.

• 생태와환경
• /
• 제40권2호
• /
• pp.346-351
• /
• 2007

Change of carabid beelte (Coleoptera, Carabidae) diversity and population structures in Woopo Wetlands (Changneung-Gun, Gyeungsangnam-Do, S. Korea) were investigated after flooding events. During the investigation period, 11 species belonging to five genera were identified. Dolichus halensis halensis(Schaller), Chlaenius (Ch.) pallipes Gebler, Ch. (Ilaenchus) naeviger Morawitz, and Pheropsophus (Stenaptinus) jessoensis Morawitz were the predominant species in Woopo Wetlands. Floods occurred twice, August and September in 2004. After the flooding events, species diversity decreased and species assemblage structures changed dramatically. Changes of the diversity and species assemblage structures were more evident in August then in September, because water level was much higher and inundation period was longer than September. A non-linear patterning algorithm of the Self-Organizing Map (SOM) was applied to discover the relationship between flooding events and carabid beetles community dynamics. Although abundance of the majority species decreased after the flooding events, that of the predominant species increased. Further detailed studies on species distribution and emigration patterns will likely bring a new insight in understanding of the adaptation mechanism of carabid beetles in wetlands.