• Journal of the Korean Data and Information Science Society
• /
• 제9권2호
• /
• pp.345-355
• /
• 1998

다층전방향 신경망을 학습시키기 위해 역전파 알고리즘이 널리 사용되고 있으나 이 알고리즘은 긴 훈련시간, 극소점 문제, 이상치에 민감하다는 단점을 가지고 있다. 한편 실제문제에서는 많은 경우에 자료에 과대오차와 이상치가 포함되게 된다. 따라서 과대 오차에 민감하지 않고, 이상치의 영향을 최소화시키는 로버스트 역전파 알고리즘의 필요성이 대두되었다. 본 논문에서는 기존의 두종류의 로버스트 역전파 알고리즘을 이론적으로 비교하고 비선형 회귀 함수추정과 문자인식과 같은 패턴인식 문제에 적용하여 실험결과를 분석한다. 그리고 향후 연구과제로 신경망 학습을 위해 베이지안 기법의 사용을 제안한다.

• Structural Engineering and Mechanics
• /
• 제21권2호
• /
• pp.205-220
• /
• 2005

This paper presents the possibilities of adapting artificial neural networks (ANNs) to predict the dimensionless parameters related to the maximum contact pressures of an elasticity problem. The plane symmetric double receding contact problem for a rigid stamp and two elastic strips having different elastic constants and heights is considered. The external load is applied to the upper elastic strip by means of a rigid stamp and the lower elastic strip is bonded to a rigid support. The problem is solved under the assumptions that the contact between two elastic strips also between the rigid stamp and the upper elastic strip are frictionless, the effect of gravity force is neglected and only compressive normal tractions can be transmitted through the interfaces. A three layered ANN with backpropagation (BP) algorithm is utilized for prediction of the dimensionless parameters related to the maximum contact pressures. Training and testing patterns are formed by using the theory of elasticity with integral transformation technique. ANN predictions and theoretical solutions are compared and seen that ANN predictions are quite close to the theoretical solutions. It is demonstrated that ANNs is a suitable numerical tool and if properly used, can reduce time consumed.

• 대한기계학회논문집A
• /
• 제24권3호
• /
• pp.634-644
• /
• 2000

Fatigue crack growth and life is estimated by various fracture mechanical parameters but affected by load, material and environment. Fatigue character of component without surface notch cannot be e valuated by above-mentioned parameters due to microstructure of in-service material. Single fracture mechanical parameter or nondestructive parameter cannot predict fatigue damage in arbitrary boundary condition but multiple fracture mechanical parameters or nondestructive parameters can Fatigue crack growth modelling with three point representation scheme uses this merit but has limit on real-time monitoring. Therefore, this study shows fatigue damage model using backpropagatior. neural networks on the basis of X-ray half breadth ratio B/$B_o$ fractal dimension $D_f$ and fracture mechanical parameters can predict fatigue crack growth rate da/dN and cycle ratioN/$N_f$ at the same time within engineering estimated mean error(5%).

• 인터넷정보학회논문지
• /
• 제6권5호
• /
• pp.45-56
• /
• 2005

컴퓨터 네트워크의 확대 및 인터넷 이용의 급격한 증가에 따라 네트워크 서비스 품질의 보장과 네트워크의 관리가 어려울 뿐만 아니라 네트워크 보안의 취약성으로 인하여 해킹 및 정보유출 등의 위협에 노출되어 있다. 특히 시스템 침입의 보안 위협에 대한 능동적인 대처 및 침입 이후에 동일하거나 유사한 유형의 사건 발생에 대해 실시간에 대응하는 것이 중요하므로 침임 탐지 시스템에 대한 많은 연구가 진행되고 있다. 본 논문에서는 시스템 호출을 이용하여 이상 침입 탐지 시스템의 성능을 향상시키기 위해, 특징 선택과 가변 길이 데이터를 고정 길이 학습 패턴으로 변환 생성하는 문제를 해결하기 위한 사운덱스 알고리즘을 적용한 신경망 학습을 통하여 이상 침입 탐지의 연구를 하고자 한다. 즉, 가변 길이의 순차적인 시스템 호출 데이터를 사운덱스 알고리즘에 의한 고정 길이의 행위 패턴을 생성하여 역전파 알고리즘에 의해 신경망 학습을 수행하였다. 역전파 신경망 기법을 UNM의 Sendmail Data Set을 이용하여 시스템 호출의 이상 탐지에 적용하여 성능을 검증하였다.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2006년도 춘계종합학술대회
• /
• pp.115-119
• /
• 2006

채널은 무선망에 있어서 한정된 주요 자원 중의 하나이다. 다양한 채널 관리 기법들이 제시되어 왔으며, 최근 들어 가드채널의 최적화 문제가 부각되고 있다. 본 논문에서는 신경망을 이용한 지능적인 채널 관리 기법을 제안한다. 신경망의 학습 데이터 생성과 성능분석을 위하여 SRN(Stochastic Reward Net) 채널 할당 모델이 개발된다. 제안된 기법에서 신경망은 지도학습 방법인 역전파 알고리즘을 이용하여 최적의 가드채널 값 g를 계산하도록 학습한다. 학습된 신경망을 이용하여 최적의 g를 계산하고, 이를 SRN모델에서 구해진 결과와 비교한다. 실험결과는 신경망에서 구한 가드채널 수와 SRN 모델로부터 구한 가드채널 수의 상대적 차이가 없음을 보여준다.

• 한국공작기계학회논문집
• /
• 제11권2호
• /
• pp.103-111
• /
• 2002

Out of all metal-cutting process, the hole-making process is the most widely used. It is estimated to be more than 30% of the total metal-cutting process. It is therefore desirable to monitor and detect drill wear during the hole-drilling process. One important aspect in controlling the drilling process is monitoring drill wear status. There are two systems, Basic system and Online system, to detect the drill wear. Basic system comprised of spindle rotational speed, feed rates, thrust torque and flank wear measured by tool microscope. Outline system comprised of spindle rotational speed feed rates, AE signal, flank wear area measured by computer vision, On-line monitoring system does not need to stop the process to inspect drill wear. Backpropagation neural networks (BPNs) were used for on-line detection of drill wear. The output was the drill wear state which was either usable or failure. This paper deals with an on-line drill wear monitoring system to fit the detection of the abnormal tool state.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 하계종합학술대회 논문집
• /
• pp.1017-1020
• /
• 1999

Recently, a new learning algorithm for multilayer neural networks has been proposed 〔1〕. In the new learning algorithm, each output neuron is considered as a function of weights and the weights are adjusted so that the output neurons produce desired outputs. And the adjustment is accomplished by taking gradients. However, the gradient computation was performed numerically, resulting in a long computation time. In this paper, we derive the all necessary equations so that the gradient computation is performed analytically, resulting in a much faster learning time comparable to the backpropagation. Since the weight adjustments are accomplished by summing the gradients of the output neurons, we will call the new learning algorithm “multi-gradient.” Experiments show that the multi-gradient consistently outperforms the backpropagation.

• 한국지능시스템학회논문지
• /
• 제9권5호
• /
• pp.463-471
• /
• 1999

This paper presents a vector controlled induction motor is implemented by neural networks system compared with PI controller for the speed control. The design employed the training strategy with Neural Network Controller(NNC) and Neural Network Emulator(NNE) for speed. In order to update the weights of the controller First of all Emulator updates its parameters by identifying the motor input and output next it supplies the error path to the output stage of the controller using backpropagation algorithm, As Controller produces an adequate output to the system due to neural networks learning capability Vector controlled induction motor characteristics actual motor speed with based on neural network system follows the reference speed better than that of linear PI speed controller.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.625-628
• /
• 1997

In this paper the new neural network and sliding mode suspension controller is proposed. That neural network is error self-recurrent neural network. For fast on-line learning, this paper use recursive least squares method. A new neural networks converges considerably faster than the backpropagation algorithm and has advantages of being less affected by the poor initial weights and learning rate. The controller for suspension systems is designed according to sliding mode technique based on new proposed neural network.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 학술대회 논문집 정보 및 제어부문
• /
• pp.349-351
• /
• 2004

The classical dynamic backpropagation learning algorithm has the problems of learning speed and the determine of learning parameter. The Extend Kalman Filter(EKF) is used effectively for a state estimation method for a non linear dynamic system. This paper presents a learning algorithm using Dual Extended Kalman Filter(DEKF) for Fully Recurrent Neural Network(FRNN). This DEKF learning algorithm gives the minimum variance estimate of the weights and the hidden outputs. The proposed DEKF learning algorithm is applied to the system identification of a nonlinear SISO system and compared with dynamic backpropagation learning algorithm.