• 대한전기학회논문지:전기기기및에너지변환시스템부문B
• /
• 제52권9호
• /
• pp.448-453
• /
• 2003

In this paper the controlled closing algorithm of the circuit breaker for isolated capacitor banks is proposed. The characteristics of circuit breakers such as RDDS(rate of decrease of dielectric strength) and a mechanical operating tolerance should be also taken into account to locate contact touch instants around voltage zero. The analysis results on the voltages across circuit breaker terminals upon closing operation play an important roles to describe the suggested method. The simulations carried out in order to verify the performance of this method show that the closing instants obtained from the proposed method provide a good suppression effect on the closing transients for both single and back-to-back capacitor banks.

• 한국생산제조학회지
• /
• 제22권5호
• /
• pp.859-864
• /
• 2013

Creation of a stress relief groove is a fairly simple yet high-performance method. During the application of this method, it is important to consider the location and size of the groove in order to achieve better performance. Consequently, this research proposes an approach for optimizing the application of the stress relief groove method to a press-fitted assembly. In a boss design, the position and diameter of the groove are configured as design variables and the design of experiments is applied. Based on this information, a 3D model is built and analyzed using the finite element analysis software ABAQUS. Meta-models are created using back-propagation neural networks. Then, deterministic optimization results obtained from a genetic algorithm are compared with the results of the finite element analysis. The temperature sensitivity of the optimized model is analyzed, and finally, reliability-based design optimization is conducted for enhancing the design quality.

• 한국음향학회지
• /
• 제13권6호
• /
• pp.5-14
• /
• 1994

이 논문은 오차역전달(error back-propagation) 알고리듬을 갖는 다층구조 퍼셉트런(Multi-Layered Perceptron)을 사용하여 우리말 단어음성을 화자종속으로 기계 인식하는 실험에 관한 연구 결과다. 대상단어는 시외 자동전화 지역번호표에서 임의로 선택한 50개 지역명이며, 이 중 43개는 2음절로 구성되어있고 나머지 7개는 3음절이다. 단어를 음소나 음절별로 분리(segmentation)하지 않고, 단어의 각 부분에서 골고루 추출된 특징성분을 신경망에 입력하는 방법을 사용했다. 그렇게 함으로써 발음지속시간에 관계없는 결과를 얻을 수 있으며, 이 때 사용된 특징 성분은 선형예측분석으로 구해진 PARCOR계수다. 전체학습과 구분학습의 비교, 프레임 갯수와 PARCOR차수에 대한 인식률의 의존도, 중간층 뉴런의 갯수에 대한 인식률의 변동, 그리고 출력층 뉴런의 구성 방법에 따른 비교 등 4가지 실험을 통하여 가장 최량의 조건을 찾아보고자 하였다. 이 연구를 발전시킨다면 실시간의 화자독립 소규모어휘 음성인식이 가능해질 것으로 보인다.

• Journal of Power Electronics
• /
• 제17권5호
• /
• pp.1223-1230
• /
• 2017

The back emf harmonics of a permanent magnet (PM) synchronous motor is a major source of torque ripple. For torque control applications including column fitted MDPS (motor driven power steering) systems, it is essential to reduce the mechanical vibrations due to torque ripples at low speeds. In this paper, a torque ripple reduction algorithm for interior PM synchronous motors is proposed. The harmonic currents that cancel the $6^{th}$ order torque harmonic are added to the nominal dq currents for MTPA (maximum torque per ampere) operation. The compensated harmonic currents are derived from flux linkage harmonics based on a FFT analysis of the back emf harmonics. Simulation and experimental results verify that the $6^{th}$ order torque harmonic and THD of the torque ripple are reduced by compensating the dq harmonic currents.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 ISIS 2003
• /
• pp.315-317
• /
• 2003

Neural oscillator is applied in oscillatory systems (Analysis of image information, Voice recognition. Etc...). If we apply established EBPA(Error back Propagation Algorithm) to oscillatory system, we are difficult to presume complicated input's patterns. Therefore, it requires more data at training, and approximation of convergent speed is difficult. In this paper, I studied the neural oscillator as synchronized states with appropriate phase relation between neurons and recognized the Korean alphabet using Neural Oscillator Phase model Synchronization.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1997년도 추계학술대회 학술발표 논문집
• /
• pp.275-279
• /
• 1997

최근 인공지능연구에서는 기호즈의와 커넥션니즘이 독립적으로 연구되어 왔으나 차츰 융합의 필요성이 절실히 요구되고 있다. 본 연구에서는 먼저 기호주의의 일부분인 고전논리를 확장한 다치논리와 커넥션니즘의 기본부분인 신경회로망을 융합한 다치신경망을 구성하고, BP에 기반을 둔 학습 MVL 네트워크를 이용하여 해석한다. 본 논문에서는 이러한 구성 및 해석방법을 확장하여 비고전적인 다치신경회로망을 구성하는 방법을 제안한다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 춘계학술대회 학술발표 논문집
• /
• pp.49-52
• /
• 1998

Neural Network has good performance in pattern classification, control and many other fields by learning ability. However, there is effective rule or systematic approach to determine optimal structure. In this paper, we propose a new method to find optimal structure of feed-forward multi-layer neural network as a kind of pruning method. That eliminating redundant elements of neural network. To find redundant elements we analysis error and weight changing with Rough Set Theory, in condition of executing back-propagation leaning algorithm.

• 설비공학논문집
• /
• 제12권6호
• /
• pp.616-622
• /
• 2000

This paper concerns the application of smoke and fire spread to road tunnel fire problems. When a road tunnel is on fire. a fire protection system of road tunnel have to offer an adequate escape space to human. Therefore, this study carried out a simulation for predicting a spreading path of smoke and fire. The evolution of the flow field is simulated with the low Reynolds number k-$\varepsilon$ turbulent model and SIMPLE algorithm based on the finite volume method.

• 한국경영과학회지
• /
• 제30권4호
• /
• pp.99-115
• /
• 2005

For an analysis of the performance of a computer system, the minimum cycle time method has been widely used. The minimum cycle time method is a mathematical technique with which we can find the minimum duration time needed to fire all the transitions at least once and coming back to the Initial marking in a timed net. A timed net is a modified version of a Petri net where a transition is associated with a delay time. In the real world, an event is associated with a probability of occurrence. However, a timed net is not equipped with any facility of specifying probability of event occurrence. Therefore, the minimum cycle time method applied on a timed net can easily overlook probabilities of occurrences of events and yield a wrong result. We are proposing 'Timed Net with Probabilities of Choices' where a transition can be associated with both delay time and a probability of occurrence. We also introduce an algorithm for minimum cycle time analysis on a 'Timed Net with Probabilities of Choices'. As an example of application, we are performing an analysis of the location based service system using 'Timed Net with Probabilities of Choices'.

• 대한기계학회논문집A
• /
• 제25권12호
• /
• pp.2062-2069
• /
• 2001

It has been established that a crack has an important effect on the dynamic behavior of a structure. This effect depends mainly on the location and depth of the crack. To identify the location and depth of a crack in a structure, a method is presented in this paper which uses synthetic artificial intelligent technique, that is, Adaptive-Network-based Fuzzy Inference System(ANFIS) solved via hybrid learning algorithm(the back-propagation gradient descent and the least-squares method) are used to learn the input(the location and depth of a crack)-output(the structural eigenfrequencies) relation of the structural system. With this ANFIS and a continuous evolutionary algorithm(CEA), it is possible to formulate the inverse problem. CEAs based on genetic algorithms work efficiently for continuous search space optimization problems like a parameter identification problem. With this ANFIS, CEAs are used to identify the crack location and depth minimizing the difference from the measured frequencies. We have tried this new idea on a simple beam structure and the results are promising.