• 대한전자공학회논문지
• /
• 제26권4호
• /
• pp.35-44
• /
• 1989

본 논문에서는 모델과 플랜트의 완전 모델 추종 제어 조건이나 명확한 매개변수 추정자를 필요로 하지않는 직접적응제어방식을 이용한 직류전동기의 정밀속도 제어장치가 제안되었다. MC-68000 microcomputor를 이용한 실제적 시스템뿐만 아니라 컴퓨터 시뮬레이션 역시 이산적인 직접적응제어 알고리듬으로 구성되었으며, 전동기의 출력이 부하에 의한 외란과 매개변수의 변동에서도 기준 입력을 잘 추동함을 알 수 있었다. 적용된 알고리듬은 소프트웨어와 하드웨어적으로 모두 간단하며 효과적인 특성을 가지고 있다.

• 전자공학회논문지C
• /
• 제35C권3호
• /
• pp.79-89
• /
• 1998

The Error Back-Propagation(EBP) algorithm is widely applied to train a multi-layer perceptron, which is a neural network model frequently used to solve complex problems such as pattern recognition, adaptive control, and global optimization. However, the EBP is basically a gradient descent method, which may get stuck in a local minimum, leading to failure in finding the globally optimal solution. Moreover, a multi-layer perceptron suffers from locking a systematic determination of the network structure appropriate for a given problem. It is usually the case to determine the number of hidden nodes by trial and error. In this paper, we propose a new algorithm to efficiently train a multi-layer perceptron. OUr algorithm uses stochastic perturbation in the weight space to effectively escape from local minima in multi-layer perceptron learning. Stochastic perturbation probabilistically re-initializes weights associated with hidden nodes to escape a local minimum if the probabilistically re-initializes weights associated with hidden nodes to escape a local minimum if the EGP learning gets stuck to it. Addition of new hidden nodes also can be viewed asa special case of stochastic perturbation. Using stochastic perturbation we can solve the local minima problem and the network structure design in a unified way. The results of our experiments with several benchmark test problems including theparity problem, the two-spirals problem, andthe credit-screening data show that our algorithm is very efficient.

• 전자공학회논문지
• /
• 제54권2호
• /
• pp.61-67
• /
• 2017

이 논문에서는 협대역과 낮은 전송속도의 제약을 가진 수중의 다양한 환경에 적응적인 MAC 기법을 제안하였다. 이미 발표된 Underwater Packet Flow Control (UPFC) 기법은 전송회수와 전송시간을 이용한 세 가지 전송타입을 제안하여 일반전송, 블럭전송, 병렬전송이 가능하도록 설계되었으나 기존 머지전송을 보완하여 최대전송가능 페이로드를 기준으로 효율적인 큐잉방법에 대한 기법 Effective Queuing-UPFC (EQ-UPFC)을 제안하였다. 제안한 기법의 전송효율의 검증을 위해 수학적 모델을 제시하였으며 시뮬레이션을 수행하여 전송회수 및 전송량 대비 효율성의 증가를 검증하였다.

• 전기전자학회논문지
• /
• 제19권2호
• /
• pp.254-259
• /
• 2015

CABAC은 문맥 기반 적응적 이진 산술 부호화 방식으로, 이전까지 부호화 된 심볼들의 정보를 이용하여 확률을 업데이트하여 부호화 효율을 높이는 기법이다. 문맥 모델러는 통계적 상관성을 고려하여 심볼에 따라 확률 모델을 설계하는 CABAC의 핵심 블록으로서, 본 논문에서는 문맥 모델러의 효율적인 하드웨어 아키텍쳐를 제안한다. Verilog HDL로 기술되어 0.18 um 공정으로 설계된 문맥 모델러는 메모리를 포함하여 29,832개의 게이트로 이루어져 있으며, 최대 동작속도는 200 MHz, 최대 처리율은 200 Mbin/s이다.

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제17권3호
• /
• pp.224-235
• /
• 2017

Purpose: Reliability is the most important factor to detect defects as wind turbines are deployed in large blades. The methods of detecting defects are various, such as non-destructive inspection and thermal imaging inspection. We propose the phased array ultrasonic testing method of non-destructive testing. Methods: We propose the active pressure mechanism for wind power blade. The phase array ultrasonic inspection method is used for fault detection inner blade surface. Controlled pressure of mechanism with respect to z-axis is important for guarantee the result of phase array ultrasonic inspection. The model based control and proposed mechanism are utilized for overall system stability and effectiveness of system. Result: The result of proposed pressure mechanism B is more stable than A. Convergence speed is also faster than A. Conclusion: We confirmed the performance of the proposed constant pressure mechanism through experiments. Non-destructive testing was applied to the specimen to confirm the reliability of detecting defects.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 추계학술대회 학술발표 논문집
• /
• pp.361-367
• /
• 1998

The main two applications of the Genetic Algorithms(GA) are the optimization and the machine learning. Machine Learning has two objectives that make the complex system learn its environment and produce the proper output of a system. The machine learning using the Genetic Algorithms is called GA machine learning or genetic-based machine learning (GBML). The machine learning is different from the optimization problems in finding the rule set. In optimization problems, the population of GA should converge into the best individual because optimization problems, the population of GA should converge into the best individual because their objective is the production of the individual near the optimal solution. On the contrary, the machine learning systems need to find the set of cooperative rules. There are two methods in GBML, Michigan method and Pittsburgh method. The former is that each rule is expressed with a string, the latter is that the set of rules is coded into a string. Th classifier system of Holland is the representative model of the Michigan method. The classifier systems arrange the strength of classifiers of classifier list using the message list. In this method, the real time process and on-line learning is possible because a set of rule is adjusted on-line. A classifier system has three major components: Performance system, apportionment of credit system, rule discovery system. In this paper, we solve the food search problem with the learning and evolution of an artificial ant using the learning classifier system.

• 한국통신학회논문지
• /
• 제33권5C호
• /
• pp.371-377
• /
• 2008

본 논문은 비전에 기반한 사람의 양팔 제스처의 모델링과 인식에 관한 연구이다. 우리는 양팔 제스처 인식을 위한 특징점의 추출에서부터 제스처의 분류에 이르는 전체적 틀을 제안하였다. 먼저, 양팔 제스처의 모델링을 위해 색채 기반의 양손 추적 방법을 제안하였고, 추출된 양손의 궤적 정보를 효과적으로 선택하게 하는 제스처 구(Phrase) 분석법을 제시하였다. 선택된 특징 점들의 시퀀스(sequence) 들로 이루어진 훈련 데이터들의 최대 공통부열(Longest Common Subsequence) 정보를 이용하여 제스처를 모델링하고 이에 따른 유사도 척도를 제안하였다. 제안된 방법론을 공항 등에서 이용하는 항공기 유도 수신호에 적용하였고, 실험을 통해 제안된 방법론의 효율성과 인식성능을 보였다.

• Industrial Engineering and Management Systems
• /
• 제11권4호
• /
• pp.310-330
• /
• 2012

Scheduling is an important tool for a manufacturing system, where it can have a major impact on the productivity of a production process. In manufacturing systems, the purpose of scheduling is to minimize the production time and costs, by assigning a production facility when to make, with which staff, and on which equipment. Production scheduling aims to maximize the efficiency of the operation and reduce the costs. In order to find an optimal solution to manufacturing scheduling problems, it attempts to solve complex combinatorial optimization problems. Unfortunately, most of them fall into the class of NP-hard combinatorial problems. Genetic algorithm (GA) is one of the generic population-based metaheuristic optimization algorithms and the best one for finding a satisfactory solution in an acceptable time for the NP-hard scheduling problems. GA is the most popular type of evolutionary algorithm. In this survey paper, we address firstly multiobjective hybrid GA combined with adaptive fuzzy logic controller which gives fitness assignment mechanism and performance measures for solving multiple objective optimization problems, and four crucial issues in the manufacturing scheduling including a mathematical model, GA-based solution method and case study in flexible job-shop scheduling problem (fJSP), automatic guided vehicle (AGV) dispatching models in flexible manufacturing system (FMS) combined with priority-based GA, recent advanced planning and scheduling (APS) models and integrated systems for manufacturing.

• 전력전자학회논문지
• /
• 제10권2호
• /
• pp.155-161
• /
• 2005

본 논문에서는 MRAS 퍼지제어를 이용한 회전자 시정수 추정 기법을 제안한다. 회전자 자속을 추정하는 방법은 기준모델과 적응 회전자 모델을 이용한다 이 두 모델은 MRAS의 형태로 구성되며 두 모델의 오차를 영으로 근접하게 제어한다. 두 모델의 파라미터가 정확하면 동일한 결과를 얻는다. 그러나 회전자 시정수의 추정이 정확하게 이루어지지 않으면 두 회전자 자속의 추정은 서로 다른 각도를 가지게 된다. 두 모델의 오차와 오차 변화분을 입력으로 퍼지 제어기를 이용하여 회전자 시정수를 추정한다.

• Journal of Communications and Networks
• /
• 제14권5호
• /
• pp.501-509
• /
• 2012

This paper proposes a joint timing synchronization, channel estimation, and data detection for the impulse radio ultra-wideband systems. The proposed timing synchronizer consists of coarse and fine timing estimation. The synchronizer discovers synchronization points in two stages and performs adaptive threshold based on the maximum pulse averaging and maximum (MAX-PA) method for more precise synchronization. Then, iterative channel estimation is performed based on the discovered synchronization points, and data are detected using the selective rake (S-RAKE) detector employing maximal ratio combining. The proposed synchronizer produces two signals-the start signal for channel estimation and the start signal for start frame delimiter (SFD) detection that detects the packet synchronization signal. With the proposed synchronization, channel estimation, and SFD detection, an S-RAKE receiver with binary pulse position modulation binary phase-shift keying modulation was constructed. In addition, an IEEE 802.15.4a channel model was used for performance comparison. The comparison results show that the constructed receiver yields high performance close to perfect synchronization.