• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.05a
• /
• pp.289-292
• /
• 2002

2계 선형 상미분방정식의 경계치 문제는 보통 해를 구하고자 하는 구간의 양 끝점에서 도함수의 값을 임의로 선정한 후 각 점에서 초기치 문제의 해를 구한 다음 적절한 1차 결합을 이용하여 구하게 된다. 이 경우 초기값과 도함수 값을 사용한 반복연산이 수반되며 따라서 오차의 누적이 불가피 하게 된다. 이 논문에서는 이같은 오차의 누적을 피할 뿐 아니라 3차 Spline 함수를 사용함으로써 오차가 O( $h^2$)인 해를 구하는 방법에 대하여 기술한다 두 개의 경계조건과 근사값을 구하고자 하는 점에서의 함수 값을 "If x is $B_{i}$, then f is $C_{i}$"와 같은 Fuzzy Rule들로 변형하고 주어진 미분방정식을 상수 $C_{i}$들의 관계식으로 변형하여 해를 구하였다. 산출된 결과로부터의 보간 연산은 Fuzzy System사용에 의하여 대체되었다. 이상의 방법으로 산출한 해의 근사오차가 O( $h^2$).임을 증명하였으며 3개의 예제에 대한 계산결과를 4계 Runge-Kutta 방법에 의한 해와 비교하여 기술하였다였다였다였다

• The Korean Journal of Applied Statistics
• /
• v.17 no.2
• /
• pp.219-228
• /
• 2004

In this paper we construct prediction intervals for nonlinear time series models using the bootstrap. We compare these prediction intervals to traditional asymptotic prediction intervals using quasi-score estimation function and M-quasi-score estimating function comprising bounded functions. Simulation results show that the bootstrap method leads to improved accuracy. The accuracy of the bootstrap is empirically demonstrated with the consumer price index.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.5
• /
• pp.1119-1131
• /
• 2016

The Neyman-Scott Rectangular Pulses Model (NSRPM) is mainly used to construct hourly rainfall series. This model uses a modest number of parameters to represent the rainfall processes and underlying physical phenomena, such as the arrival of storms or rain cells. In NSRPM, the method of moments has often been used because it is difficult to know the distribution of rainfall intensity. Recently, approximated likelihood function for NSRPM has been introduced. In this paper, we propose a hierarchical model for applying a spatial structure to the NSRPM parameters using the approximated likelihood function. The proposed method is applied to summer hourly precipitation data observed at 59 weather stations (Korea Meteorological Administration) from 1973 to 2011.

• Journal of Internet of Things and Convergence
• /
• v.6 no.4
• /
• pp.49-57
• /
• 2020

In this paper, I provided an approximation function Li*2,10(x) using logarithm integral for the counting function π*2,10(x) of consecutive deca primes. Several personal computers and Mathematica were used to validate the approximation function Li*2,10(x). I found the real value of π*2,10(x) and approximate value of Li*2,10(x) for various x ≤ 1011. By the result of theses calculations, most of the error rates are margins of error of 0.005%. Also, I proved that the sum C2,10(∞) of reciprocals of all primes with difference 10 between primes is finite. To find C2,10(∞), I computed the sum C2,10(x) of reciprocals of all consecutive deca primes for various x ≤ 1011 and I estimate that C2,10(∞) probably lies in the range C2,10(∞)=0.4176±2.1×10-3.

• Solar Energy
• /
• v.15 no.2
• /
• pp.25-37
• /
• 1995

This paper presents an approximate analytical solution to one-dimensional model of the charging process for stratified thermal storage tanks, in which variation of the inlet temperature as well as the momemtum-induced mixing is taken into accout. The mixing is incorporated into the model as a constant-depth perfectly mixed layer above the plug flow region. Based on the superposition principle, the variable inlet temperature is approximated by a number of step functions. Temperature distributions for the thermocline corresponding to three types of interfacial condition arr successfully derived in terms of well-defined functions, so that a linear combination of them constitutes the final solution. Validity and utility of this work is examined through the comparison of the approximate solution with an exact solution available for the case of linearly increasing inlet temperature. With increasing the number of steps, the present solution asymptotically approaches to the exact one. Even with a limited number of steps, the present results favorably agree with those by the exact solution for a wide range of the mixing depth. Also, it is revealed that fewer steps are needed for meaningful predictions as the mixing. depth becomes larger.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.6
• /
• pp.47-52
• /
• 2002

The wavelet functions are originated from scaling functions and can be used as activation function in the hidden node of the network by deciding two parameters such as scale and center. In this paper, we would like to propose the mixed structure. When we compose the WNN using wavelet functions, we propose to set a single scale function as a node function together. The properties of the proposed structure is that while one scale function approximates the target function roughly, the other wavelet functions approximate it finely. During the determination of the parameters, the wavelet functions can be determined by the global search algorithm such as genetic algorithm to be suitable for the suggested problem. Finally, we use the back-propagation algorithm in the learning of the weights.

• Proceedings of the Korea Contents Association Conference
• /
• 2005.05a
• /
• pp.341-345
• /
• 2005

In this paper, we propose an algorithm for the approximation of FIR filters by IIR filters. The algorithm is based on a concept of the balanced model reduction. Head-related transfer functions(HRTFs) of dummy-head are approximated by 32-order IIR filters. The binaural sounds using the approximated HRTFs are reproduced by headphone, and serves as a cue of sound image localization. Experiment of sound image are carried out for 10 participants with computer simulation and DSP board respectively. The results of the experiments show that the localization using the approximated HRTFs by IIR filters is the same accuracy as the case of FIR filters that simulate the HRTFs.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.16 no.4 s.95
• /
• pp.418-426
• /
• 2005

Based upon three level approximation and the steepest descent path(SDP) method, we consider an optimum choice of approximation path for derivation of new class of closed-flrm Green's functions which can lead to the analytic evaluation of MoM(Method of Moment) matrix elements. It is observed that the present method can give more accurate evaluation of the spatial Green's functions than the previous method, even without the advance investigation of the spectral functions, over a wide frequency range. In order to check the validity of the present method, some numerical results are presented.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.23 no.7
• /
• pp.822-830
• /
• 2012

Clutters are all unwanted radar returns to affect on detection of targets. Radar clutter is characterized by amplitude distributions, spectrum, etc. Clutter is modelled with considering these kinds of characteristics. In this paper, a Weibull distribution function approximated by uniform distribution function is suggested. Weibull distribution function is used to model the various clutters. This paper shows that the data generated by the approximated solution of Weibull distribution function satisfy the Weibull probability density function. This paper shows that the data generation time of approximated Weibull distribution function solution is reduced by 20 % compared with the generation time of original Weibull probability density function.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.2
• /
• pp.115-122
• /
• 2012

We present here a new class of activation functions for neural networks, which herein will be called CosGauss function. This function is a cosine-modulated gaussian function. In contrast to the sigmoidal-, hyperbolic tangent- and gaussian activation functions, more ridges can be obtained by the CosGauss function. It will be proven that this function can be used to aproximate polynomials and step functions. The CosGauss function was tested with a Cascade-Correlation-Network of the multilayer structure on the Tic-Tac-Toe game and iris plants problems, and results are compared with those obtained with other activation functions.