• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제8권1호
• /
• 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
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• 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.

• 대한기계학회논문집A
• /
• 제24권11호
• /
• pp.2742-2748
• /
• 2000

In this study, the quantitative evaluation was performed by examination for gatigue crack origin in two prepared ductile irons specimens which have different microstructures using rotary bending fatigue tester, Series A has mixed microstructure. ferrite plus pearlite, and series B has bainitic microstructure. Obtained main results are as follows. The scatterings of fatigue strength were observed on σ(sub)α-Ν diagram of both specimen series, and it is observed that scatterings of series A were more serious. It is reasonable to evaluate the size of mesocrack range by means of √A(원문참조), where A means the area of mesocrack range including globular graphite nodule. As a result of reconsideration for the fatigue data by introduction of K(sub)α-Ν diagram, the scattrings of fatigue life were, remarkably. reduced. Therefore, it is more reasonable to evaluate of mesocrack range on fatigue life by parameter K(sub)α rather than σ(sub)α.

• 대한조선학회논문집
• /
• 제42권2호
• /
• pp.129-135
• /
• 2005

At present, fatigue design of welded structures is primarily based on a nominal stress or hot spot stress approach with a series of classified weld S-N curves. Although well accepted by major industries, the nominal stress based fatigue design approach is cumbersome in terms of securing a series of S-N curves corresponding to each class of joint types and loading modes. The hot spot stress based fatigue design has a difficulty of finding a proper stress through the global model, the midium size model, and the detail model of ship structure. Also, it is difficult to link proper displacements within three different mesh size models. Recently, the structural stress is proposed as a mesh-size insensitive structural stress definition that gives a stress state at weld toe with relatively large mesh size. However, this method requires an experimental validation in obtaining the fatigue strength of weldments. Therefore, in this study, a series of experiment is performed for various sizes of weldments.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 추계학술대회 논문집 전력기술부문
• /
• pp.137-139
• /
• 2003

Micro-Sources arc emerging as a providing environmentally friendly and reliable and effective solution. Micro-Source converter DC voltage of prime mover to AC voltage using voltage source inverter for interjecting with AC grid system. This paper studies optimal series inductance size for interjecting with AC grid system.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 하계학술대회 논문집 A
• /
• pp.300-302
• /
• 1996

In this paper a new series-parallel compensated uninterruptible power supply is proposed. Its series compensator shapes input current to sinusoid. The power handled by series compensator is only a quarter of ratings. And parallel compensator delivers sinusoidal voltage to nonlinear load. The parallel compensator is backedup with battery. This system has capabilities of power line conditioner and backup power with reduced size.

• 한국기계가공학회지
• /
• 제11권1호
• /
• pp.82-87
• /
• 2012

For different ferrite-pearlite matrix structure, contain more than 90% spheroidal ratio of graphite, GCD 45-3, GCD 50, GCD 60 series and 70%, 80%, 90% spheroidal ratio of graphite, GCD 40, GCD 45-1, GCD 45-2 series, this paper has carried out rotary bending fatigue test, estimated maximum and mean size of spheroidal graphite, investigated correlation. It was concluded as follows. (1) Fatigue limit in $10^7$cycles and numbers of spheroidal graphite per 1$mm^2$ was linear relation. (2) projection area of graphite can be used to predict fatigue limit of Ductile Cast Iron. The Statistical distribution of extreme values of projection area of defects may be used as a guideline for the control of inclusion size in the steelmaking processes.

• 전력전자학회논문지
• /
• 제13권2호
• /
• pp.110-118
• /
• 2008

기존의 저궤도 인공위성 전력계 시스템에 사용된 직렬 구조, 병렬 구조는 공전 주기 동안 두 번의 전력변환 과정을 거쳐 시스템의 효율을 감소시킨다. 직-병렬 구조는 한 번의 전력변환 과정을 거쳐 시스템의 효율을 향상시키지만 레귤레이터가 추가됨으로써 시스템의 비용과 무게, 크기를 증가시킨다. 본 논문에서는 레귤레이터의 추가 없이 시스템의 효율을 향상시키는 단순화된 직-병렬 구조를 제안한다. 인공위성이 지구를 공전하는 동안 상황에 따라 변화하는 제안한 구조의 동작을 네 가지 모드로 분류하고, 각 모드마다 제안한 시스템을 안정적인 동작을 확인하기 위하여 대신호 분석을 수행하였다. 실험을 통하여 기존의 직렬구조와 제안한 구조의 효율을 비교하였다. 제안한 구조의 안정적인 동작을 검증하기 위해 200W급 태양전지와 TMS320F2812 DSP로 제어되는 100W급 전력조절기 두 모듈을 병렬로 구성하여 실험하였다.

• Advances in nano research
• /
• 제12권5호
• /
• pp.467-482
• /
• 2022

In the current work, static and free torsional vibration of functionally graded (FG) nanorods are investigated using Fourier sine series. The boundary conditions are described by the two elastic torsional springs at the ends. The distribution of functionally graded material is considered using a power-law rule. The systems of equations of the mechanical response of nanorods subjected to deformable boundary conditions are achieved by using the modified couple stress theory (MCST) and taking the effects of torsional springs into account. The idea of the study is to construct an eigen value problem involving the torsional spring parameters with small scale parameter and functionally graded index. This article investigates the size dependent free torsional vibration based on the MCST of functionally graded nano/micro rods with deformable boundary conditions using a Fourier sine series solution for the first time. The eigen value problem is constructed using the Stokes' transform to deformable boundary conditions and also the convergence and accuracy of the present methodology are discussed in various numerical examples. The small size coefficient influence on the free torsional vibration characteristics is studied from the point of different parameters for both deformable and rigid boundary conditions. It shows that the torsional vibrational response of functionally graded nanorods are effected by geometry, small size effects, boundary conditions and material composition. Furthermore, for all deformable boundary conditions in the event of nano-sized FG nanorods, the incrementing of the small size parameters leads to increas the torsional frequencies.

• 대한산업공학회지
• /
• 제28권2호
• /
• pp.171-177
• /
• 2002

In this paper, we propose a new data fusion method to improve the performance of individual prediction models for time series data. Individual models used are ARIMA and neural network and their results are combined based on the weight reflecting the inverse of EWMA of squared prediction error of each individual model. Monte Carlo simulation is used to identify the situation where the proposed approach can take a vintage point over typical fusion methods which utilize MSE for weight. Study results indicate the following: EWMA performs better than MSE fusion when the data size is large with a relatively big amplitude, which is often observed in intra-cranial pressure data. Additionally, EWMA turns out to be a best choice among MSE fusion and the two individual prediction models when the data size is large with relatively small random noises, often appearing in tax revenue data.