• 한국통신학회논문지
• /
• 제21권2호
• /
• pp.541-551
• /
• 1996

본 연구에서는 인공 신경 회로망을 이용하여 LCD 변수들의 값을 동적으로 변화시킴으로써 보다 개선된 전압 적정유지율을 얻을 수 있는 실시간 ULTC 제어전략이 개발된다. 제안된 전략에서는 수전전압의 변화에 따른 주변압기 송출전압 변화를 인식하는 ANNs, 그리고 ANNs로부터의 전압레벨과 배전선로들의 시간대별 변화패턴을 인식하여, ULTC의 정정치를 동적으로 결정하는 ANNg를 도입함으로서 보다 개선된 전압보상능력을 얻을 수 있도록 하였다. 개발된 제어전략의 성능을 평가하기 위해서 8개의 피더로 구성되는 시험 배전계통에 대해서 부하가 불규칙적으로 변화하였을때, 그리고 부하가 일정한 시간대별 패턴으로 변화하였을때의 ULTC의 전압 보상 전략이 모의된다. 인공 신경회로망은 Fortran 언어로 구현되며 시험계통에 대한 성능평가에서 유용한 결과를 입증하였다.

• 한국지능시스템학회논문지
• /
• 제26권5호
• /
• pp.409-415
• /
• 2016

최근 사물인터넷 기반의 재해예방 기술이 개발되고 있다. 본 논문에서는 사물인터넷기반의 공동주택용 자율전기안전관리 기술 개발을 위하여 부하 전류 파형을 FFT와 MFCC를 이용하여 신호변환 후 신경회로망 모델에 적용하여 정확도가 개선된 전기 부하 패턴분류 시스템을 제안한다. 오실로스코프와 CT를 이용하여 측정한 전기 부하의 전류 파형을 FFT 알고리즘을 적용한 후 신경회로망을 이용하여 단일부하패턴 분류 실험을 하였다. 본 연구를 통하여 부하의 특성을 파악함으로서 고장에 대해 보다 신속하고 정확하게 대처할 수 있을 것으로 예측된다.

• 대한기계학회논문집
• /
• 제17권2호
• /
• pp.393-406
• /
• 1993

In resistance spot welding process, size of molten nuggest have been utilized to assess the integrity of the weld quality. However real-time monitoring of the nugget size is an extremely difficult problem. This paper describes the design of an artificial neural networks(ANN) estimator to predict the nugget size for on-line use of weld quality monitoring. The main task of the ANN estimator is to realize the mapping characteristics from the sampled dynamic resistance signal to the actual negget size through training. The structure of the ANN estimator including the number of hidden layers and nodes in a layer is determined by an estimation error analysis. A series of welding experiments are performed to assess the performance of the ANN estimator. The results are quite promissing in that real-time estimation of the invisible nugget size can be achieved by analyzing the dynamic resistance signal without any conventional destructive testing of welds.

• Journal of Biosystems Engineering
• /
• 제23권6호
• /
• pp.571-582
• /
• 1998

The PLS(Partial Least Square) and ANN(Artificial Neural Network) were introduced to develop the soluble solids content prediction model of apples which is followed by making a subsequent selection of photosensor. For the optimal PLS model, number of factors needed for spectrum analysis were increased until the convergence of prediction residual error sum of squares. Analysis has shown that even part of the overall wavelength with no pretreatment may turn out better performing. The best PLS model was found in the 800 to 1,100nm wavelength region without pretreatment of second derivation, having $R^2$=0.9236, bias= -0.0198bx, SEP=0.2527bx for unknown samples. On the other hand, for the ANN model the second derivation led to higher performance. On partial range of 800 to 1,100nm wavelengh region, prediction model with second derivation for unknown samples reached $R^2$=0.9177, SEP=0.2903bx in contrast to $R^2$=0.7507, SEP =0.4622bx without pretreatment.

• 대한기계학회논문집A
• /
• 제24권1호
• /
• pp.52-61
• /
• 2000

The role of quantitative nondestructive evaluation of defects is becoming more important to assure the reliability and the safety of structure, which can eventually be used for residual life evaluation of structure on the basis of fracture mechanics approach. Although ultrasonic technique is one of the most widely used techniques for application of practical field test among the various nondestructive evaluation technique, there are still some problems to be solved in effective extraction and classification of ultrasonic signal from their noisy ultrasonic waveforms. Therefore, crack size determination through a neural network based on the back-propagation algorithm using back-scattered ultrasonic signals is established in this study. For this purpose, aluminum plate containing vertical or inclined surface breaking crack with different crack length was used to receive the back-scattered ultrasonic signals by pulse echo method. Some features extracted from these signals and sizes of cracks were used to train neural network and the neural network's output of the crack size are compared with the true answer.

• 한국유체기계학회 논문집
• /
• 제5권4호
• /
• pp.47-53
• /
• 2002

The steam generator feedwater flow-rate in a nuclear power plant was estimated by means of artificial neural networks with the wavelet analysis for enhanced information extraction. The fouling of venturi meters, used for steam generator feedwater flow-rate in pressurized water reactors, may result in unnecessary plant power derating. The back-propagation network was used to generate models of signals for a pressurized water reactor Multiple-input, single-output hetero-associative networks were used for evaluating the feedwater flow rate as a function of a set of related variables. The wavelet was used as a low pass filter eliminating the noise from the raw signals. The results have shown that possible fouling of venturi can be detected by neural networks, and the feedwater flow-rate can be predicted as an alternative to existing methods. The research has also indicated that the decomposition of signals by wavelet transform is a powerful approach to signal analysis for denoising.

• 한국농공학회지
• /
• 제36권1호
• /
• pp.95-102
• /
• 1994

Since parameters influencing the indoor environment of livestock building interrelate so complicatedly, it is of great difficulty to identify the exact cause of environmental problems in a livestock building. Therefore, the approaches for the problem solving based on experience not numerical calculation will be helpful to the management of livestock building This study was attempt to develop the decision supporting system to diagnose environmen- tal problems in a livestock building based on an expert system and a neural network. HClips$^3$), attaching the Hangeul user interface to Clips which is known as a powerful shell for develop- ing expert system, was used. The multilayer perceptron consisting of 4 layers including back propagation learning algorithm was adpoted, which was rapidly converged within the allowable range at 50,000 learning sweeps. The expert system and neural network seemed to work well for this specific application, providing proper suggestions for some environmental problems: particularly, the neural net- work trained by an environmental problem and its corresponding answer with certainty factor, produced the same results as those by expert system.

• 한국정보통신학회논문지
• /
• 제3권3호
• /
• pp.553-558
• /
• 1999

환자의 상태 및 심장 질환 등의 진단에 있어 매우 중요한 정보신호는 심전도 신호이며, 많은 잡음이 혼입되어 있기 때문에 잡음 신호의 필터링이 매우 어렵고 잘못된 신호처리는 심전도 신호의 왜곡을 가져올 수 있다. 심전도 신호의 잡음을 필터링하기 위해 기존의 방법은 다 단계 형태로 필터를 구성하여 처리하기 때문에 신호처리 구조가 복잡하고 연산 량이 많아 처리속도가 느려진다. 이러한 문제를 개선하기 위해 인공지능의 한 기법인 RBF 신경회로망을 이용하여 간단한 구조로 심전도 신호의 필터링 방법을 제안하고, 실험한 결과 우수한 성능을 얻었다.

• 한국정밀공학회지
• /
• 제15권7호
• /
• pp.149-159
• /
• 1998

The importance of soundness and safety evaluation in weld zone using by the ultrasonic wave has been recently increased rapidly because of the collapses of huge structures and safety accidents. Especially, the ultrasonic method that has been often used for a major non-destructive testing(NDT) technique in many engineering fields plays an important role as a volume test method. Hence, the defecting any defects of weld Bone in austenitic stainless steel type 304 using by ultrasonic wave and neural network is explored in this paper. In order to detect defects, a distance amplitude curve on standard scan sensitivity and preliminary scan sensitivity represented of the relation between ultrasonic probe, instrument, and materials was drawn based on a quantitative standard. Also, a total of 93％ of defect types by testing 30 defect patterns after organizing neural network system, which is learned with an accuracy of 99％, based on ultrasonic evaluation is distinguished in order to classify defects such as holes or notches in experimental results. Thus, the proposed ultrasonic wave and neural network is useful for defect detection and Ultrasonic Non-Destructive Evaluation(UNDE) of weld zone in austenitic stainless steel 304.

• 생물환경조절학회지
• /
• 제4권2호
• /
• pp.246-252
• /
• 1995

작물, 가축, 농산물을 학문의 대상으로 하는 농학은 기상, 토양 등과 같은 자연 현상으로부터 필요한 데이터를 획득하여 이용한다. 그러나, 이들 데이터는 많은 환경 요인의 영향을 받아 그 거동이 매우 복잡한 비선형적 현상을 나타내는 것이 대부분이다. 따라서, 실험을 통해 획득된 데이터의 처리 및 모형화 등을 위해 기존의 수학적, 통계적 방법을 이용하는 경우에 많은 어려움을 겪게 된다. 이에 최근에는 신경회로망 및 퍼지 이론 등과 같은 인공 지능 기법을 이용하여 이러한 문제점을 해결하기 위한 연구가 활발히 진행되고 있다. 본 강좌에서는 복잡한 비선형 특성 특히 임의적 거동을 보이는 자연 현상을 기술하기 위해 최근에 대두되고 있는 혼돈 이론에 대한 소개를 하고자 한다.(중략)