• Communications for Statistical Applications and Methods
• /
• 제16권6호
• /
• pp.891-901
• /
• 2009

이 논문에서는 GARCH 모형에서 가정한 오차향의 분포에 근접하도록 자료를 변환하고 변환된 자료를 이용하여 모수와 예측구간을 구한 후 다시 역변환을 통해 원래의 척도에서의 VaR을 계산하는 방법에 대해 알아본다. KOSPI와 KOSDAQ 수익률을 이동시키며 VaR을 계산하고 이들 VaR의 포함확률을 계산하여 병목수준에 얼마나 근접하는지를 알아봄으로써 변환-역변환 방법과 변환을 적용하지 않는 방법의 결과를 비교해 본다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2007년도 학술발표회 논문집
• /
• pp.58-66
• /
• 2007

This paper introduces the new generic dynamic neuro-fuzzy local modeling system (DNFLMS) that is based on a dynamic Takagi-Sugeno (TS) type fuzzy inference system for complex dynamic hydrological modeling tasks. The proposed DNFLMS applies a local generalization principle and an one-pass training procedure by using the evolving clustering method to create and update fuzzy local models dynamically and the extended Kalman filtering learning algorithm to optimize the parameters of the consequence part of fuzzy local models. The proposed DNFLMS is applied to develop the inference model to forecast the flow of Waikoropupu Springs, located in the Takaka Valley, South Island, New Zealand, and the influence of the operation of the 32 Megawatts Cobb hydropower station on springs flow. It is demonstrated that the proposed DNFLMS is superior in terms of model accuracy, model complexity, and computational efficiency when compared with a multi-layer perceptron trained with the back propagation learning algorithm and well-known adaptive neural-fuzzy inference system, both of which adopt global generalization.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 하계학술대회 논문집 B
• /
• pp.627-629
• /
• 1997

In this paper, fuzzy logic system equipped with the back-propagation training algorithm as identifiers for nonlinear dynamic systems is described. To improve its performance, Jacob's delta-bar -delta rule is adapted in adjusting stepsize ${\alpha}$, and only y and ${\alpha}$ updating algorithm is suggested. In identifying and predicting the chaotic time series, suggested method is better than Li-Xin Wang's method,[1]

• 설비공학논문집
• /
• 제3권3호
• /
• pp.206-214
• /
• 1991

Flow field in a hard disk drive has been predicted numerically. Theoretical model was constructed based on a commercially available hard disk drive with 40 Mega byte capacity. Since the gap between disk tip and shroud is not homogeneous in real hard disk drive, three kinds of gap size have been tested as computational model. The discussion has been made on the circumferential velocity, radial velocity, and pressure fields. As a result, the average shear stress on the disk surface was reduced as the gap size decreased. This means that the shroud should be designed compactly to reduce power consumption of the spindle motor.

• 제어로봇시스템학회논문지
• /
• 제13권6호
• /
• pp.589-595
• /
• 2007

This paper presents a new method on controller design of brushless dc motors. In such drives the current ripples are generated by motor inductance in stator windings and the back EMF. To suppress the current ripples the current controller is generally used. To minimize the size and the cost of the drives it is desirable to control motors without the current controller and the current sensing circuits. To estimate the motor CUlTent it is modeled by a neural network that is contigured as an output-error dynamic system. The identified model is essentially a one step ahead prediction structure in which past inputs and outputs are used to calculate the current output. Using the model, a state feedback controller to compensate the effects of disturbance has been designed. The controller is implemented by a 16-bit microprocessor and the effectiveness of the proposed control method is verified through experiments.

• 전기학회논문지
• /
• 제58권11호
• /
• pp.2101-2106
• /
• 2009

The power system state estimation and prediction are very important for operation. Because that accidents of the Power system are the cause that many devices and etc are damaged. Currently, almost every power systems have 2nd,3rd back-upsystem for prevention of accident. But prevention of accident by miss-operation, due to operator or miss data, has not acounter plan. Because, we need to estimate the power system for correcting miss data and preventing miss operation by operator. We suggest algorithm for integrity of power system network data.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2005년도 제36회 하계학술대회 논문집 B
• /
• pp.1156-1158
• /
• 2005

This paper deals with the prediction of characteristic for permanent magnet linear synchronous motor(PMLSM). The open-circuit field distribution is predicted using a two-dimensional(2-D) analytical solution derivd in terms of magnetic vector potential. The slotting and skew effect is considered using the relative permeance function. and than using this result, flux linkage and back EMF is calculated. The results are validated extensively by finite element(FE) analyses and measurement.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2001년도 학술발표회 논문집(I)
• /
• pp.45-51
• /
• 2001

The daily inflow is apparently one of nonlinear and complicated phenomena. The nonlinear and complexity make it difficult to model the prediction of daily flow, but attractive to try the neural networks approach which contains inherently nonlinear schemes. The study focuses on developing the forecasting models of daily inflows to a large dam site using neural networks. In order to reduce the error caused by high or low outliers, the back propagation algorithm which is one of neural network structures is modified by combining a regression algorithm. The study indicates that continuous forecasting of a reservoir inflow in real time is possible through the use of modified neural network models. The positive effect of the modification using tole regression scheme in BP algorithm is showed in the low and high ends of inflows.

• 대한용접접합학회:학술대회논문집
• /
• 대한용접접합학회 1992년도 특별강연 및 추계학술발표 개요집
• /
• pp.91-95
• /
• 1992

The resistance spot welding process has been extensively used for joining of sheet metals, which are subject to variation of many process variables. Many qualitive analyses of sampled process variables have been attempted to predict nugget size. In this paper, dynamic resistance and electrode movement signal which is a good indicative of the nugget size was examined by introducing an artificial neural network estimator. An artificial neural feedforward network with back-propagation of error was applied for the estimation of the nugget size. The prediction by the neural network is in good agreement with the actual nugget size for resistance spot welding of galvanized steel. The results are quite promising in that the quantitative estimation of the invisible nugget size can be achieved without conventional destructive testing of welds.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2014년도 춘계학술발표대회
• /
• pp.305-308
• /
• 2014

인간은 성격에 따라 이동패턴이 변화한다고 한다. 이런 점에서 인간의 성격 데이터를 이용하면, 인간의 행동 패턴을 유추해 낼 수 있다. 우리는 실제 실험자들의 GPS데이터와 BFI성격 데이터를 수집하고. Back Propagation Network를 이용하여, 새로운 위치 데이터를 추론하는 과정을 설명하였다. 논문의 내용은 다음과 같다. 첫 번째로 BFI(Big-Five Inventory) 성격평가에 대해 설명한다. 두 번째로 GPS데이터와 성격 데이터를 실험에 적절한 형태로 변환하는 방법에 대해 언급하고, 세 번째로 변환된 데이터를 이용하여 사람의 새로운 위치 정보를 추론할 것이다. 마지막으로 해당 실험의 결과 및 분석 그리고 앞으로의 연구 방향에 대해 언급할 것이다.