• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.677-680
• /
• 2005

In numerical analysis on the thermal performance of the heat exchanger with phase change fluids, the numerical values of thermodynamic properties are needed. But the steam table should be modeled properly as the direct use of thermodynamic properties of the steam table is impossible. In this study the function approximation characteristics of neural networks was used in modeling the saturated vapor region of refrigerant R12. The neural network consists of one input layer with one node, two hidden layers with 10 and 20 nodes each and one output layer with 7 nodes. Input can be both saturation temperature and saturation pressure and two cases were examined. The proposed model gives percentage error of ${\pm}$0.005% for enthalpy and entropy, ${\pm}$0.02% for specific volume and ${\pm}$0.02% for saturation pressure and saturation temperature except several points. From this results neural network could be a powerful method in function approximation of saturated vapor region of R12.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.3
• /
• pp.297-305
• /
• 2008

Recently, kernel methods have attracted great interests in the areas of pattern classification, function approximation, and anomaly detection. The role of the kernel is particularly important in the methods such as SVM(support vector machine) and KPCA(kernel principal component analysis), for it can generalize the conventional linear machines to be capable of efficiently handling nonlinearities. This paper considers the problem of combining radar and rain gauge observations utilizing the regression approach based on the kernel-based gaussian process and support vector learning. The data-assimilation results of the considered methods are reported for the radar and rain gauge observations collected over the region covering parts of Gangwon, Kyungbuk, and Chungbuk provinces of Korea, along with performance comparison.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.4
• /
• pp.549-560
• /
• 1993

A simple closed form approximation is developed by a new approach presented in this paper for the microstrip surface dyadic Green's function which arises in the problem of an electric current point source on an infinite planar grounded dielectric substrate. This closed form approximation includes the effects of the space wave, the surface wave and their coupling within the transition region near the source, and remains accurate as near as $0.1{\pi}_1$ from the source point for a substrate thickness as large as $0.04{\pi}_1$, where, ${\pi}_1$, is the free space wavelength, This result can significantly facilitate the rigorous moment method analysis of microstrip antenna arrays on relatively this substrates of practical interest. Numerical results illustrating the accuracy of the closed form approximation are presented and CPU times associated with some mutual impedance calculations are also included.

• The Korean Journal of Applied Statistics
• /
• v.20 no.1
• /
• pp.103-115
• /
• 2007

In this paper we studied the small sample asymptotic inference for the autoregressive coefficient in AR(1) model. Based on saddlepoint approximations to the distribution of quadratic forms, we suggest a new approximation to the distribution of the estimators of the noncircular autoregressive coefficients. Simulation results show that the suggested methods are very accurate even in the small sample sizes and extreme tail area.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.6
• /
• pp.511-516
• /
• 2002

The neural networks may have problem such that the amount of calculation for the network learning goes too big according to the dimension of the dimension. To overcome this problem, the wavelet neural networks(WNN) which use the orthogonal basis function in the hidden node are proposed. One can compose wavelet functions as activation functions in the WNN by determining the scale and center of wavelet function. In this paper, when we compose the WNN using wavelet functions, we set a single scale function as a node function together. We intend that 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 for solutions suitable for the suggested problem using the genetic algorithm and finally, we use the back-propagation algorithm in the learning of the weights.

• Proceedings of the KIPE Conference
• /
• 2013.11a
• /
• pp.39-40
• /
• 2013

본 논문에서는 직접토크 제어에서 자속과 토크의 오차에 따라 결정된 전압벡터와 회전자 위치에 따른 실제 인가될 수 있는 d-q축 전압을 근사 전압함수로 근사화하여, 자속 및 토크오차와 전동기의 속도에 따라 듀티비를 결정하는 방식을 제안한다. 이러한 방식은 선택된 전압벡터가 일정한 상수 크기의 전압을 인가하는 것으로 가정된 기존의 직접토크 제어 방식에 비해 정밀한 전압 기준을 바탕으로 펄스폭의 듀티비를 결정함으로써, 동일한 스위칭 주파수 내에서 토크 및 자속오차의 크기를 감소 시킬 수 있는 장점이 있다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.10a
• /
• pp.390-396
• /
• 1998

본 논문은 퍼지 논리 제어기(FLC)의 근사화 능력과 제어 성능을 동시에 향상시키는 정확한 무게 중심(Center Of Gravity; COG) 비퍼지화기를 제안한다. 본 논문은 비퍼지화 과정이 최적 선택의 한 과정이며 비퍼지화 방법의 적절한 선택이다. 제안한 COG 비퍼지화기의 정확성은 출력 소속 함수를 여러 개의 설계 파라메터(중신, 폭, 변경자(modifier))로 나타내고 이들 설계 파라메터들을 학습과 진화의 Lamarckian 상호 적응에 의하여 갱신함으로써 얻어진다. 이러한 학습과 진화의 상호 적응은 학습하지 않는 경우 보다 빠르게 최적 COG 비퍼지화기를 얻도록 하며, 보다 넓은 범위의 탐색으로최적해를 찾을 가능성을 높여 준다. 제안한 설계 방법은 목적 함수의 가중치를 조절하여 높은 근사화 능력, 높은 제어 성능, 또는 이들간의 균형된 성능을 갖는 다양한 특정 응용형(Application-specific)COG 비퍼지화기를 제공한다. 제안한 상호적응 COG 비퍼지화기의 설계방법을 트럭 후진 주차 제어 문제에 적용하여, 각각 시스템 오차와 평균 추적 거리로 나타내어진 근사화 능력과 제어 성능을 기존의 COG 비퍼지화기와 비교한다.

• The Korean Journal of Applied Statistics
• /
• v.20 no.2
• /
• pp.257-266
• /
• 2007

It is known that Agresti-Coull approach is an effective tool for the construction of confidence intervals for various problems related to binomial proportions. However, the Agrest-Coull approach often produces a conservative confidence interval. In this note, confidence intervals based on a Bayesian approach are proposed for a linear function of independent binomial proportions. It is shown that the Bayesian confidence interval slightly outperforms the confidence interval based on Agresti-Coull approach in average sense.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.21 no.2
• /
• pp.117-126
• /
• 2009

In order to test the accuracy between the modified mild slope equation (MMSE) without evanescent modes and the plane-wave approximation (PA) of eigenfunction expansion method, various numerical results from both models are presented. In this study, analytical solutions of two models are employed, one based on the MMSE derived by Porter (2003) and the other on the scatterer method of PA by Seo (2008a). Judging from direct comparisons against existing results of rapidly varying topography, the PA model gives better predictions of the wave propagation than the MMSE model.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.861-868
• /
• 2023

One of the key technologies in providing data analysis in the deep learning while maintaining security is fully homomorphic encryption. Due to constraints in operations on fully homomorphically encrypted data, non-arithmetic functions used in deep learning must be approximated by polynomials. Until now, the degrees of approximation polynomials with composite minimax polynomials have been uniformly set across layers, which poses challenges for effective network designs on fully homomorphic encryption. This study theoretically proves that setting different degrees of approximation polynomials constructed by composite minimax polynomial in each layer does not pose any issues in the inference on convolutional neural networks.