• Journal of Platform Technology
• /
• v.6 no.4
• /
• pp.34-40
• /
• 2018

Currently, artificial neural networks perform well for a single task, but NN have the problem of forgetting previous learning by learning other kinds of tasks. This is called catastrophic forgetting. To use of artificial neural networks in general purpose this should be solved. There are many efforts to overcome catastrophic forgetting. However, even though there was a lot of effort, it did not completely overcome the catastrophic forgetting. In this paper, we propose sequential iterative learning using core concepts used in elastic weight consolidation (EWC). The experiment was performed to reproduce catastrophic forgetting phenomenon using EMNIST data set which extended MNIST, which is widely used for artificial neural network learning, and overcome it through sequential iterative learning.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.22 no.2
• /
• pp.126-132
• /
• 2010

An approach to calculating expected damage of breakwater assisted by artificial neural network was developed. Wave height in front of a breakwater was predicted by a trained artificial neural network with inputs of wave height in deep ocean and tidal level. Prediction results by the neural network can be comparable to that by professional numerical model for wave transformation. Using the wave prediction neural network, it was very easy and fast to obtain a number of significant waves at breakwater and finally analysis time for expected damage can be shortened. In addition, the effect of considering tidal level in the calculation of expected damage was revealed by comparing the expected damages with and without tidal variation. Therefore, it was pointed out that tidal variation should be considered to improve prediction accuracy.

• Journal of Korean Society of Steel Construction
• /
• v.21 no.2
• /
• pp.175-181
• /
• 2009

An intelligent control technique using a neural network is proposed for offshore structures exposed to sea-bed earthquakes. Fluid-structure interaction effect was considered in developing controller and a training algorithm for the neural network is presented. In the numerical example, the performance of the proposed neural network controller was compared with that of a passive controller and uncontrolled structures. Based on the example, it can be concluded that the proposed neuro-control scheme can be used for offshore structures with nonlinear characteristics due to its interaction with fluid.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.1
• /
• pp.89-97
• /
• 2008

In this paper, we present a method for the enhancement of marker detection correctness and marker recognition speed by using artificial neural network. Contours of objects are extracted from the input image. They are approximated to a list of line segments. Quadrangles are found with the geometrical features of the approximated line segments. They are normalized into exact squares by using the warping technique and scale transformation. Feature vectors are extracted from the square image by using principal component analysis. Artincial neural network is used to checks if the square image is a marker image or a non-marker image. After that, the type of marker is recognized by using an artificial neural network. Experimental results show that the proposed method enhances the correctness of the marker detection and recognition.

• Journal of the Korean Geotechnical Society
• /
• v.19 no.4
• /
• pp.287-298
• /
• 2003

The feasibility of using neural networks to model the complex relationship between piezocone measurements and the undrained shear strength of clays has been investigated. A three layered back propagation neural network model was developed based on actual undrained shear strengths, which were obtained from the isotrpoically and anisotrpoically consolidated triaxial compression test(CIUC and CAUC), and piezocone measurements compiled from various locations around the world. It was validated by comparing model predictions with measured values about new piezocone data, which were not previously employed during development of model. Performance of the neural network model was compared with conventional empirical method, direct correlation method, and theoretical method. It was found that the neural network model is not only capable of inferring a complex relationship between piezocone measurements and the undrained shear strength of clays but also gives a more precise and reliable undrained shear strength than theoretical and empirical approaches. Furthermore, neural network model has a possibility to be a generalized relationship between piezocone measurements and undrained shear strength over the various places and countries, while the present empirical correlations present the site specific relationship.

• Journal of the Korean GEO-environmental Society
• /
• v.13 no.8
• /
• pp.25-33
• /
• 2012

Although it is important to reflect the accurate information of the ground condition in the tunnel design, the analysis and design are conducted by limited information because it is very difficult to consider various geographies and geotechnical conditions. When the tunnel is under construction, examination of accurate safety and prediction of behavior are overcome the limits of predicting behavior by Artificial Neural Network in this study. First, construct the suitable structure after the data of field was made sure by the multi-layer back propagation, then apply with algorithm. Employ the result of measured data from database, and consider the influence factor of tunnel, like supporting pattern, RMR, Q, the types of rock, excavation length, excavation shape, excavation over, to carry out the reliable analysis through field applicability of Artificial Neural Network. After studying, using the ANN model to predict the shearing displacement, convergence displacement, underground displacement, Rock bolt output follow the excavation over of tunnel construction field, then determine the field applicability with ANN through field measured value and comparison analysis when tunnel is being constructed.

• The Journal of Engineering Geology
• /
• v.28 no.4
• /
• pp.741-753
• /
• 2018

Local drought in South Korea has recently increased interest in the efficient use of groundwater and then induces a growing need to introduce artificial recharge of groundwater that stores water in sedimentary layer. In order to evaluate the potential artificial recharge sites in the alluvial basins in Chungcheongnamdo province, an AHP (Analytical hierarchy process) model consisting of three primary and seven secondary factors was developed in this study. In the AHP model, adding candidate sites changes final evaluation score through a mathematical calculation process. By contrast ANN (Artificial neural network) model always provides an unchanged score for each candidate area. Therefore, the score can be used as a selection criterion for artificial recharge sites. It is concluded that the possibility of artificial recharge is relatively low if the score of the ANN model is less than about 1.5. Further studies and field surveys on the other regions in Korea will lead to draw out a more applicable ANN model.

• The KIPS Transactions:PartB
• /
• v.13B no.6 s.109
• /
• pp.615-624
• /
• 2006

The objective of this paper is to implement parallel multi-layer ANN(Artificial Neural Network) simulator based on the mobile agent system which is executed in parallel in the virtual parallel distributed computing environment. The Multi-Layer Neural Network is classified by training session, training data layer, node, md weight in the parallelization-level. In this study, We have developed and evaluated the simulator with which it is feasible to parallel the ANN in the training session and training data parallelization because these have relatively few network traffic. In this results, we have verified that the performance of parallelization is high about 3.3 times in the training session and training data. The great significance of this paper is that the performance of ANN's execution on virtual parallel computer is similar to that of ANN's execution on existing super-computer. Therefore, we think that the virtual parallel computer can be considerably helpful in developing the neural network because it decreases the training time which needs extra-time.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2010.05a
• /
• pp.1269-1273
• /
• 2010

이러한 강우자료의 결측값이나 오자료를 보정하는 것은 그 유역의 정확한 수문학적 특성 파악 및 안전한 수공구조물의 설계에 영향을 미치게 되므로 매우 중요하다고 할 수 있다. 최근 이러한 강우자료를 비선형적 모델인 인공신경망(Artificial Neural Network)을 이용하여 보정하는 연구가 활발히 진행되고 있다(오재우 등, 2008). 그러나 이러한 인공신경망을 적용하는 경우, 선택한 신경망 구조의 형태와 학습(training)을 위해 사용되는 자료가 전체 자료의 특성을 반영하고 있는 정도에 따라 정확도에 차이를 보인다(한광희 등, 2010). 따라서 자료보정을 위한 입력 자료의 선택은 인공신경망을 이용한 결측치 보정의 중요한 과정이다. 본 연구에서는 이러한 입력 자료의 선택을 위한 여러 가지 기법 중 입력 변수간의 상호정보량 (Mutual Information)을 이용한 방법을 적용하여 대상 결측 지점을 보정할 강우지점을 선별한 후 선택된 지점만으로 인공신경망을 구성하여 강우자료를 보정하고 주변 자료를 모두 이용한 결과와 상관성분석으로 얻어진 결과와 비교하였다.

• Journal of the Korean Geotechnical Society
• /
• v.21 no.9
• /
• pp.25-34
• /
• 2005

This paper proposes a selection methodology composed of neural network (NN) and genetic algorithm (GA) to design optimal NN structure. We combine the characteristics of GA and NN to reduce the computational complexity of artificial intelligence applications and increase the precision of NN' prediction in the design of NN structure. Genetic selection approach of design parameters of NN is introduced to obtain optimal NN structure. Analyzed results for geotechnical problems are given to evaluate the performance of the proposed hybrid methodology.