• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 1999년도 추계학술대회-지능형 정보기술과 미래조직 Information Technology and Future Organization
• /
• pp.365-373
• /
• 1999

Recently, numerous studies have demonstrated that artificial intelligence such as neural networks can be an alternative methodology for classification problems to which traditional statistical methods have long been applied. In building neural network model, the selection of independent and dependent variables should be approached with great care and should be treated as a model construction process. Irrespective of the efficiency of a learning procedure in terms of convergence, generalization and stability, the ultimate performance of the estimator will depend on the relevance of the selected input variables and the quality of the data used. Approaches developed in statistical methods such as correlation analysis and stepwise selection method are often very useful. These methods, however, may not be the optimal ones for the development of neural network models. In this paper, we propose a genetic algorithms approach to find an optimal or near optimal input variables for neural network modeling. The proposed approach is demonstrated by applications to bankruptcy prediction modeling. Our experimental results show that this approach increases overall classification accuracy rate significantly.

• 항공우주시스템공학회지
• /
• 제18권1호
• /
• pp.21-28
• /
• 2024

본 논문에서는 회전익 블레이드의 단면 구조 정보를 통해 블레이드의 단면 강성을 예측하고, 재료 정보를 이용하여 단면 강성을 예측할 수 있는 심층 신경망 기반 네트워크 예측 모델의 설계 및 적절성 검토를 수행하였다. 재료 정보를 네트워크 입력으로 갖는 예측 모델의 경우, 블레이드 단면 부재 재료의 탄성 계수를 네트워크의 입력으로 고려하여 단면 강성을 예측하도록 설계하였다. 또한, 단면 구조 정보를 네트워크 입력으로 갖는 예측 모델의 경우, 블레이드의 단면을 구성하는 단면 부재의 위치와 두께 정보를 네트워크 입력으로 고려하여 단면 강성을 예측하도록 설계하였다. 각 예측 모델은 심층신경망 구조를 기반으로 설계하였으며, 단면 해석 프로그램인 KSAC2D를 통한 단면 해석 결과를 네트워크의 훈련 및 검증 데이터로 사용하였다.

• 농업과학연구
• /
• 제47권4호
• /
• pp.1109-1122
• /
• 2020

The aim of this study was to develop a marbling classification and prediction model using small parts of sirloin images based on a deep learning algorithm, namely, a convolutional neural network (CNN). Samples were purchased from a commercial slaughterhouse in Korea, images for each grade were acquired, and the total images (n = 500) were assigned according to their grade number: 1++, 1+, 1, and both 2 & 3. The image acquisition system consists of a DSLR camera with a polarization filter to remove diffusive reflectance and two light sources (55 W). To correct the distorted original images, a radial correction algorithm was implemented. Color images of sirloins of Hanwoo (mixed with feeder cattle, steer, and calf) were divided and sub-images with image sizes of 161 × 161 were made to train the marbling prediction model. In this study, the convolutional neural network (CNN) has four convolution layers and yields prediction results in accordance with marbling grades (1++, 1+, 1, and 2&3). Every single layer uses a rectified linear unit (ReLU) function as an activation function and max-pooling is used for extracting the edge between fat and muscle and reducing the variance of the data. Prediction accuracy was measured using an accuracy and kappa coefficient from a confusion matrix. We summed the prediction of sub-images and determined the total average prediction accuracy. Training accuracy was 100% and the test accuracy was 86%, indicating comparably good performance using the CNN. This study provides classification potential for predicting the marbling grade using color images and a convolutional neural network algorithm.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 추계학술대회 논문집 학회본부 C
• /
• pp.910-912
• /
• 1998

In this paper, we purpose automatic diagnosis in online, as the fundamental study to diagnose the partial discharge mechanism and to predict the lifetime, by introduction a neural network. In the proposed method, Ire use acoustic emission sensing system and calculate a fixed quantity statistic operator by pulse number and amplitude. Using statically operators such as the center of gravity(G) and the gradient of the discharge distribute(C), we analyzed the early stage and the middle stage. the fixed quantity statistic operators are learned by a neural network. The diagnosis of insulation degradation and a lifetime prediction by the early stage time are achieved. On the basis of revealed excellent diagnosis ability through the neural network learning for the patterns during degradation, it was proved that the neural network is appropriate for degradation diagnosis and lifetime prediction in partial discharge.

• 한국공작기계학회논문집
• /
• 제16권4호
• /
• pp.79-86
• /
• 2007

The shape of excessive penetration mainly depends on welding conditions(welding current and welding voltage), and welding process(groove gap and welding speed). These conditions are the major affecting factors to width and height of back bead. In this paper, back-bead prediction and weldability estimation using artificial neural network were investigated. Results are as follows. 1) If groove gap, welding current, welding voltage and welding speed will be previously determined as a welding condition, width and height of back bead can be predicted by artificial neural network system without experimental measurement. 2) From the result applied to three weld quality levels(ISO 5817), both experimented measurement using vision sensor and predicted mean values by artificial neural network showed good agreement. 3) The width and height of back bead are proportional to groove gap, welding current and welding voltage, but welding speed. is not.

• Journal of Welding and Joining
• /
• 제21권1호
• /
• pp.80-86
• /
• 2003

This study develops a system for effective prediction of residual stresses by the backpropagation algorithm using the neural network. To achieve this goal, a series of experiments were carried out to and measured the residual stresses using the sectional method. With the experimental results, the optional control algorithms using a neural network could be developed in order to reduce the effect of the external disturbances during GMA welding processes. Then the results obtained from this study were compared between the measured and calculated results, weld guality might be controlled by the neural network based on backpropagation algorithm.. This system can not only help to understand the interaction between the process parameters and residual stress, but also improve the quantity control for welded structures.

• 한국표면공학회지
• /
• 제36권3호
• /
• pp.284-289
• /
• 2003

In this study, the prediction on the quality of tools after coating process has been investigated. Under different coating conditions, cutting resistances have been obtained and analyzed with a tool dynamometer to provide optimized coating conditions. The optimized coating conditions Lhave been computed with the most effective factors found by S/N ratio of Taguchi method. To evaluate the influence of the factors on cutting efficiency through the minimum of number of experiment times, the way of neural network design using Taguchi method has been employed.

• 전기학회논문지
• /
• 제65권12호
• /
• pp.2084-2093
• /
• 2016

It is crucial to predict water temperature for aquatic ecosystem studies and management. In this paper, we first address challenging issues in predicting water temperature in a real time manner and propose a distributed computing model to address such issues. Then, we present an Artificial Neural Network (ANN)-based water temperature prediction model developed for the Soyang River and a cyberinfrastructure system called WT-Agabus to run such prediction models in an automated and real time manner. The ANN model is designed to use only weather forecast data (air temperature and rainfall) that can be obtained by invoking the weather forecasting system at Korea Meteorological Administration (KMA) and therefore can facilitate the automated and real time water temperature prediction. This paper also demonstrates how easily and efficiently the real time prediction can be implemented with the WT-Agabus prototype system.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2008년도 춘계학술발표논문집
• /
• pp.248-250
• /
• 2008

본 논문에서는 교통사고를 예방하고 교통사고 구간 대기시간을 줄이기 위해서 신경망을 이용한 예측방법을 제안한다. 뿐만 아니라, 교통사고 예측에 있어서 신경망에 정규화하지 않은 데이터를 사용하는 방법을 제시한다. 이 방법은 신경망 훈련시 데이타의 최대 값을 추정할 필요가 없어 정규화 없이 신경망을 사용 가능하며, 신뢰성 예측 결과도 추정된 최대 값과 실제 획득될 최대 값과의 차이(추정 오차)만큼 정확해질 수 있다. 또한 비정규화 된 데이터를 사용하는 방법이 데이터의 최대값을 알고 있다고 가정한 상태의 정규화된 방법보다 예측 정확성이 좋음을 보였다. 모의실험결과 제안된 신경망 예측시스템이 신경망을 고려하지 않은 기존방법보다 교통사고 구간 대기시간을 줄일 수 있음을 입증했다. 이와 같이 검증된 예측능력을 바탕으로 사용자에게 교통상황을 실시간으로 서비스하기 위하여 센서 데이터베이스를 이용한 실시간 교통정보 예보 시스템을 제안한다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2002년도 추계학술대회 논문집
• /
• pp.646-649
• /
• 2002

It can be effect on diagnosis of hydraulic machining system to analyze working conditions with shape characteristics of wear debris in a lubricated machine. But, in order to predict and estimate working conditions, it is need to analyze the shape characteristics of wear debris and to identify. Therefor, if shape characteristics of wear debris is identified by computer image analysis and the neural network, it is possible to find the cause and effect of moving condition. In this study, wear debris in the lubricant oil are extracted by membrane filter, and the quantitative value of shape characteristics of wear debris we calculated by the digital image processing. This morphological informations are studied and identified by the artificial neural network. The purpose of this study is In apply morphological characteristics of wear debris to prediction and estimation of working condition in hydraulic driving systems.