• Proceedings of the KIEE Conference
• /
• 1997.07c
• /
• pp.897-899
• /
• 1997

The analysis of distribution line faults is essential to the proper protections of the power system. A high impedance fault does not make enough current to cause conventional protective devices. It is well known that undesirable operating conditions and certain types of faults on electric distribution feeders cannot be detected by using conventional protection system. This paper describes an algorithm using back-propagation neural network for pattern recognition and detection of high impedance faults. Fractal dimensions are estimated for distinction between random noise and chaotic behavior in the power system. The fractal dimension of the line current is also used as a indication of the high impedance fault.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.1155-1157
• /
• 1996

In this paper, the direct adaptive control using neural networks is presented for the control of chaotic nonlinear systems. The direct adaptive control method has an advantage that the additional system identification procedure is not necessary. Two direct adaptive control methods are applied to a Duffing's equation and the simulation results show the effectiveness of the controllers.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.43.1-43
• /
• 2001

We propose two control methods of the Lyapunov exponents for Jordan-type recurrent neural networks. Both the two methods are formulated by a gradient-based learning method. The first method is derived strictly from the definition of the Lyapunov exponents that are represented by the state transition of the recurrent networks. The first method can control the complete set of the exponents, called the Lyapunov spectrum, however, it is computationally expensive because of its inherent recursive way to calculate the changes of the network parameters. Also this recursive calculation causes an unstable control when, at least, one of the exponents is positive, such as the largest Lyapunov exponent in the recurrent networks with chaotic dynamics. To improve stability in the chaotic situation, we propose a non recursive formulation by approximating ...

• Fisheries and Aquatic Sciences
• /
• v.11 no.1
• /
• pp.61-69
• /
• 2008

Using empirical data of fish performance and physiological limits as well as physical stimuli and environmental data, a cod-end selectivity model based on a chaotic behavior model using the psycho-hydraulic wheel and neural-network approach was established to predict fish escape or herding responses in trawl and cod-end designs. Fish responses in the cod-end were categorized as escape or herding reactions based on their relative positions and reactions to the net wall. Fish movements were regulated by three factors: escape time, a visual looming effect, and an index of body girth-mesh size. The model was applied to haddock in a North Sea bottom trawl including frequencies of movement components, swimming speed, angular velocity, distance to net wall, and the caught-fish ratio; simulation results were similar to field observations. The ratio of retained fish in the cod-end was limited to 37-95% by optomotor coefficient values of 0.3-1.0 and to 13-67% by looming coefficient values of 0.1-1.0. The selectivity curves generated by this model were sensitive to changes in mesh size, towing speed, mesh type, and mesh shape.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2000.10a
• /
• pp.37-39
• /
• 2000

The purpose of this paper is to present a neural network model based on change-point detection for the prediction of the U.S. Treasury Securities. Interest rates have been studied by a number of researchers since they strongly affect other economic and financial parameters. Contrary to other chaotic financial data, the movement of interest rates has a series of change points due to the monetary policy of the U.S. government. The basic concept of this proposed model is to obtain intervals divided by change points, to identify them as change-point groups, and to use them in interest rates forecasting. The proposed model consists of three stages. The first stage is to detect successive change points in the interest rates dataset. The second stage is to forecast the change-point group with the backpropagation neural network (BPN). The final stage is to forecast the output with BPN. This study then examines the predictability of the integrated neural network model for interest rates forecasting using change-point detection.

• The Journal of the Korea Contents Association
• /
• v.6 no.2
• /
• pp.34-43
• /
• 2006

This paper shows modeling methods for VBR video trace. It is well known that VBR video trace is characterized as longterm correlated and highly intermittent burst data. To analyze this, we attempt to model it using neural network with auxiliary linear structures derived from residual threshold. For testing purpose, we generate VBR video trace from chaotic nonlinear function combined with the geometric random noise. The modeling result of the generated data shows that the attempted method represents more accurately than the traditional neural network. However, we also found that combining hRU to the attempted modeling method can yield a closer agreement to statistical features of the generated data than the attempted modeling method alone.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1997.10a
• /
• pp.181-184
• /
• 1997

Practical applications of learning systems usually involve complex domains exhibiting nonlinear behavior and dilution by noise. Consequently, an intelligent system must be able to adapt to nonlinear processes as well as probabilistic phenomena. An important class of application for a knowledge based systems in prediction: forecasting the future trajectory of a process as well as the consequences of any decision made by e system. This paper examines the robustness of data mining tools under varying levels of noise while predicting nonlinear processes in the form of chaotic behavior. The evaluated models include the perceptron neural network using backpropagation (BPN), the recurrent neural network (RNN) and case based reasoning (CBR). The concepts are crystallized through a case study in predicting a Henon process in the presence of various patterns of noise.

• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.103-123
• /
• 1999

Input filtering as a preprocessing method is so much crucial to get good performance in time series forecasting. There are a few preprocessing methods (i.e. ARMA outputs as time domain filters, and Fourier transform or wavelet transform as time-frequency domain filters) for handling time series. Specially, the time-frequency domain filters describe the fractal structure of financial markets better than the time domain filters due to theoretically additional frequency information. Therefore, we, first of all, try to describe and analyze specially some issues on the effectiveness of different filtering methods from viewpoint of the performance of a neural network based forecasting. And then we discuss about neural network model architecture issues, for example, what type of neural network learning architecture is selected for our time series forecasting, and what input size should be applied to a model. In this study an input selection problem is limited to a size selection of the lagged input variables. To solve this problem, we simulate on analyzing and comparing a few neural networks having different model architecture and also use an embedding dimension measure as chaotic time series analysis or nonlinear dynamic analysis to reduce the dimensionality (i.e. the size of time delayed input variables) of the models. Throughout our study, experiments for integration methods of joint time-frequency analysis and neural network techniques are applied to a case study of daily Korean won / U. S dollar exchange returns and finally we suggest an integration framework for future research from our experimental results.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.8 no.1
• /
• pp.31-36
• /
• 2008

This paper presents adaptive learning data of evolvable neural networks (ENNs) for time series prediction of nonlinear dynamic systems. ENNs are a special class of neural networks that adopt the concept of biological evolution as a mechanism of adaptation or learning. ENNs can adapt to an environment as well as changes in the enviromuent. ENNs used in this paper are L-system and DNA coding based ENNs. The ENNs adopt the evolution of simultaneous network architecture and weights using indirect encoding. In general just previous data are used for training the predictor that predicts future data. However the characteristics of data and appropriate size of learning data are usually unknown. Therefore we propose adaptive change of learning data size to predict the future data effectively. In order to verify the effectiveness of our scheme, we apply it to chaotic time series predictions of Mackey-Glass data.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.920-924
• /
• 2005

This paper presents adaptive learning data of evolvable neural networks (ENNs) for time series prediction of nonlinear dynamic systems. ENNs are a special class of neural networks that adopt the concept of biological evolution as a mechanism of adaptation or learning. ENNs can adapt to an environment as well as changes in the environment. ENNs used in this paper are L-system and DNA coding based ENNs. The ENNs adopt the evolution of simultaneous network architecture and weights using indirect encoding. In general just previous data are used for training the predictor that predicts future data. However the characteristics of data and appropriate size of learning data are usually unknown. Therefore we propose adaptive change of learning data size to predict the future data effectively. In order to verify the effectiveness of our scheme, we apply it to chaotic time series predictions of Mackey-Glass data.