• Annual Conference of KIPS
• /
• 2008.11a
• /
• pp.27-29
• /
• 2008

시스템 분석에 주로 사용하는 자료 중에는 비선형 자료와 시계열 등이 있다. 이들 자료는 그 함축적인 관계가 매우 복잡하여 전통적인 통계분석 도구로 분석하는데 어려움이 많다. 본 연구에서는 현실 세계에서 다양하게 나타나는 복잡성을 다루기 위하여 하이브리드 진화 신경망 모델링 접근 방법으로 자료를 모형화 하고 이를 통한 학습의 적합도를 살펴본다. 비선형 자료 등을 모형화하기 위한 학습은 역전파 신경망 기법을 이용한다. 학습의 효율을 높이기 의해서 격자감소 학습 알고리즘과 함께 이용하는 유전자 알고리즘은 네트워크 구조를 최적화 시킬 수 있는 초기가중값을 이용한 전역 최소값을 찾는데 이용한다. 학습 결과를 통해 제안된 하이브리드형 접근방법의 학습이 보다 효율적임을 살펴보기 위하여 유전자 알고리즘으로 최적화된 신경망 학습 알고리즘을 비선형 모의자료의 학습에 적용하여 보았다.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1990.10a
• /
• pp.143-147
• /
• 1990

자동차는 주행시에 노면가진으로 인하여 진동을 받게 되는 데 이러한 진동 은 운전자의 승차감을 저하시킬 뿐만 아니라 화물과 차량 부품을 파손시키 는 원인이 되기도 한다. 노면가진력은 주로 현가장치를 통하여 차량에 전달 되어 차량의 진동을 유발하게 되므로 승차감을 향상시키려면 노면가진을 효 과적으로 차단할 수 있도록 현가장치를 설계하여야 한다. 이를 위하여 본 논 문에서는 차량을 현가장치의 형식에 따라 3가지 종류로 구분한 다음 각각을 12자유도로 모델링하여 노면이 불규칙(random)한 도로를 주행할 때의 승용 차의 진동을 해석하고, 현가스프링의 강성과 쇼크 압소바의 감쇠 값이 자체 수직 가속도 유효치(rms value)와 현가장치 유효행정(rms stroke length)에 미치는 영향을 분석하였다. 차량의 운동방정식은 현가장치의 비선형 특성을 고려하였기 때문에 미소 각 운동의 가정에도 불구하고 비선형 방정식이 된 다. 운동 방정식의 해는 4차 Runge-Kutta 적분법을 사용하여 수치해석으로 구하였다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.05a
• /
• pp.329-332
• /
• 2003

본 논문에서는 새로이 제안된 자기 구조화하는(Self-structuring) 새로운 Radial-Basis Function Network(RBFN)에 대해서 실험적인 검증을 했다. 이 자기 구조화하는 새로운 RBFN은 기존의 RBFN과 비교해서 여러 장점이 있다. Lyapunov 이론에 기초해서 새로운 학습 규칙을 선정하였기 때문에 시스템의 안정도를 보장할 수 있다. 그리고, 자기 구조화의 과정 즉, 생성과 병합을 통해 은닉층에서 적정수의 뉴런을 결정할 수 있다. 기존의 RBFN과 성능을 비교하기 위하여, 실제 비선형 시스템인 2축 암로봇에 대해 실험한 결과를 보였다. 결과적으로, 우리는 실험결과를 통해 자기 구조화하는 RBFN의 효율적인 구조와 시스템에 대한 안정도를 보장함을 볼 수 있다.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.7 no.1
• /
• pp.33-45
• /
• 1995

This paper as a sequel to Kang et al. (1994) describes the development of a three-dimensional nonlinear Galerkin-FEM model. Nonlinear advective terms have been incorporated in a manner used by Lardner and Song (1992), that is, using velocities at given nodes computed with linear Galerkin FEM model. The Proposed model is computationally more efficient than previous nonlinear Galerkin models developed by Owen (1980) and Davies (1980) because the model uses a linear shape function as a basis and, furthermore, the similarity transform technique developed by Kang (1994). Two experiments have been carried out to examine effects of nonlinear terms. One is an experiment of wind-driven current in a rectangular basin (Heaps' basin) and the other is an experiment concerning eddy generation behind a jetty with specified downstream and upstream open boundary conditions. The computed Pattern was found to be in good agreement qualitatively with previous model experiments by Stelling (1984).

• Journal of Korean Association for Spatial Structures
• /
• v.8 no.5
• /
• pp.47-58
• /
• 2008

Spatial structure is an appropriate shape that resists external force only with in-plane force by reducing the influence of bending moment, and it maximizes the effectiveness of structural system. With this character of the spatial structure, generally long span is used. As a result, large deflection is accompanied from the general frame. the structure is apt to result in a large deflection even though this structure experiences a small displacement in absence. Usually, nonlinear analysis in numerical analysis means geometric nonlinearity and material nonlinearity and complex nonlinearity analysis considers both of them. In this study, nonlinear equation of equilibrium considering geometric nonlinearity as per finite element method was applied and also considered the material nonlinearity using the relation of stress-strain in element. It is applied to find unstable result for tracing load-deflection curve in the numerical analysis tech. especially Arc-length method, and result of the analysis was studied by ABAQUS a general purpose of the finite element program. It is found that the present analysis predicts accurate nonlinear behavior of plane and space truss.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.20 no.5
• /
• pp.424-430
• /
• 2000

In order to develop an ultrasonic evaluation method for properties degradation of high temperature materials, a number of Cr-Mo-V steel samples were heat-treated and their damage mechanism was examined. Ultrasonic parameters such as velocity, attenuation, and more recently developed nonlinear acoustic parameter were measured. The nonlinear acoustic parameter was found to be most sensitive to material degradation mainly attributed to the precipitation of impurities in grain boundaries. When compared to the electrical resistivity results, the nonlinear parameters showed similar behavior. There existed a relatively good correlation between the nonlinear parameter and the fracture appearance transition temperature (FATT) obtained by Charpy V-notch impact test. Based on the relationship between the FATT and the fracture toughness ($K_{IC}$), correlation between the nonlinear parameter and $K_{IC}$ was established.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.8 no.6 s.40
• /
• pp.73-80
• /
• 2004

A nonlinear time-domain system identification algorithm using measured acceleration data is developed for structural damage assessment. To take account of nonlinear behavior of structural systems, an output error between measured and computed acceleration increments has been defined and a constrained nonlinear optimization problem is solved for optimal structural parameters. The algorithm estimates time-varying properties of stiffness and damping parameters. Nonlinear response of restoring force of a structural system is recovered by using the estimated time-varying structural properties and computed displacement by Newmark-$\beta$ method. In the recovery, no pre-defined model for inelastic behavior has been assumed. In developing the algorithm, noise and incomplete measurement in space and state have been considered. To examine the developed algorithm, numerical simulation and laboratory experimental studies on a three-story shear building have been carried out.

• Journal of Sensor Science and Technology
• /
• v.7 no.5
• /
• pp.334-341
• /
• 1998

In this paper, we present a new interpolation scheme for image enhancement using nonlinear operator. In general, interpolation techniques are based on linear operators which are essentially lowpass filters, hence, they tend to blur fine details in the original image. In our approach, the operator itself balances the strength of its sharpening and noise suppressing components according to the properties of the input image data.

• Journal of Korea Water Resources Association
• /
• v.30 no.3
• /
• pp.279-291
• /
• 1997

This study is to evaluate the accuracy and practicality of the existing four conceptual models, two linear models of Clark and Nash model and two nonlinear models of Laruenson and WBN model, and to select an appropriate model to simulate the rainfall-runoff process in a given catchment. The variability of parameters for linear models is generally larger than that of nonlinear models. The errors in peak discharge are similar among the four conceptual models buy the errors in time to peak are quite different. Nonlinear models produce better results for time distribution than linear models. A comparison of the conceptual models to predict overall hydrograph using Friedman two-way analysis of variance by rank test indicates that nonlinear models are slightly better than linear models.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.6 no.6
• /
• pp.7-15
• /
• 2002

In this paper, the nonlinear parameter estimation method using the estimated hysteresis of each structural members was studied for the purpose of efficient seismic damage prediction and estimation of MDOF nonlinear structural model in the shaking table test. The hysteresis of each structural members can be obtained by the conversion of measured response histories into relative motions of each structural members and member forces. These hysteresis can be used to evaluate various kinds of damage indices of each structural members. The MDOF nonlinear structural model for further analysis(re-analysis) can be easily reconstructed using estimated nonlinear structural parameters of each structural members. To demonstrate the proposed techniques, several numerical and experimental example analyses are carried out. The results indicate that the proposed method can be very useful to assess local seismic damages of structures.