• Journal of Electrical Engineering and information Science
• /
• v.3 no.2
• /
• pp.131-138
• /
• 1998

We derive a back-propagation learning algorithm of fuzzy neural networks using fuzzy operations, which preserves the shapes of fuzzy numbers, in order to utilize fuzzy if-then rules as well as numerical data in the learning of neural networks for classification problems and for fuzzy control problems. By introducing the shape preseving fuzzy operation into a neural network, the proposed network simplifies fuzzy arithmetic operations of fuzzy numbers with exact result in learning the network. And we illustrate our approach by computer simulations on numerical examples.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.3
• /
• pp.575-587
• /
• 2011

We investigate some statistical properties of several dierence-based error variance estimators in nonparametric regression model. Most of existing dierence-based methods are developed under asymptotical properties. Our focus is on the exact form of mean and variance for the lag-k dierence-based estimator and the second-order dierence-based estimator in a nite sample size. Our approach can be extended to Tong's estimator (2005) and be helpful to obtain optimal k.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.107.6-107
• /
• 2001

Genetic alogorithm(GA) is a well-known global optimization algorithm. However, as the searching bounds grow wider., performance of local optimization deteriorates. In this paper, we propose a hybrid algorithm which integrates the gradient algorithm and GA so as to reinforce the performance of local optimization. We apply this algorithm to the system identification of second order RLC circuit. Identification results show that the proposed algorithm gets the better and robust performance to find the exact values of RLC elements.

• 제어로봇시스템학회:학술대회논문집
• /
• 1988.10a
• /
• pp.32-36
• /
• 1988

Generally, for nonlinear control of robotic system, the method of computed torque or inverse dynamics is frequently used. In this case, exact knowledge of the system parameters is required, however. This paper addresses the problem of nonlinear control when the parameter of system is varied. The approach is based upon decoupled model following. As an example, control of a three degree of freedom manipulator arm under mass variation is simulated.

• Journal of applied mathematics & informatics
• /
• v.33 no.1_2
• /
• pp.185-191
• /
• 2015

This paper considers a discrete-time buffer system with session-based arrivals, an infinite storage capacity and one unreliable output line. There are multiple different types of sessions and the output line is governed by a finite state Markov chain. Based on a generating functions approach, we obtain an exact expression for the mean buffer content.

• Journal of Advanced Marine Engineering and Technology
• /
• v.14 no.4
• /
• pp.67-71
• /
• 1990

The traditional finite element method applied to dynamic problems employs shape functions which are based on a static displacement assumption. The more exact approach uses frequency-dependent shape functions and frequency-dependent mass and stiffness matrices. Such matrices are developed for a torsional vibration of shaft element. Numerical examples are presented for a cantilever beam.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2002.11a
• /
• pp.382.1-382
• /
• 2002

This paper presents a new approach integrated Case-Based Reasoning with Self- Organization Map(SOM) in diagnosis systems. The causes of faults are obtained by case-base trained from SOM. When the vibration problem of rotating machinery occurs, this provides an exact diagnosis method that shows the fault cause of vibration problem. In order to verify the performance of algorithm, we applied it to diagnose the fault cause of the electric motor.

• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.34D no.7
• /
• pp.1-7
• /
• 1997

The physical optical solution to the E-polarized diffraction by a composite wedge consisting of perfect conductor and lossless dielectric cannot sayisfy not only the boundary conditions at the wedge interfaces but also the edge condition at the wedge tip. Its diffraction coefficients are extended outside the composite wedge to become the exact solution to the perfectly conducting wedge as its relative dielectric constant increases to infinite or decreases to 1. It is assured that the extended diffraction coefficients approach zero in the arificially complementary region of the composite wedge.

• Journal of the Korean Statistical Society
• /
• v.36 no.4
• /
• pp.479-488
• /
• 2007

The elliptic random field is an extension to the Gaussian random field. We proved a theorem which characterizes the elliptic random field. We proposed a heuristic approach to derive an approximation to the distribution of the size of one connected component of its excursion set above a high threshold. We used this approximation to approximate the distribution of the largest cluster size. We used simulation to compare the approximation with the exact distribution.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.7 no.1
• /
• pp.25-31
• /
• 2003

The concepts of stability of algorithms for solving the symmetric and generalized symmetric-definite eigenproblems are discussed. An algorithm for solving the symmetric eigenproblem $Ax={\lambda}x$ is stable if the computed solution z is the exact solution of some slightly perturbed system $(A+E)z={\lambda}z$. We use both normwise approach and componentwise way of measuring the size of the perturbations in data. If E preserves symmetry we say that an algorithm is strongly stable (in a normwise or componentwise sense, respectively). The relations between the stability and strong stability are investigated for some classes of matrices.