• E2M - 전기 전자와 첨단 소재
• /
• 제11권11호
• /
• pp.1-8
• /
• 1998

This paper presents an arificial neuro-fuzzy technique based prtial discharge (PD) pattern classifier to power system application. This may require a complicated analysis method employ -ing an experts system due to very complex progressing discharge form under exter-nal stress. After referring briefly to the developments of artificical neural network based PD measurements, the paper outlines how the introduction of new emerging technology has resulted in the design of a number of PD diagnostic systems for practical applicaton of residual lifetime prediction. The appropriate PD data base structure and selection of learning data size of PD pattern based on fractal dimentsional and 3-D PD-normalization, extraction of relevant characteristic fea-ture of PD recognition are discussed. Some practical aspects encountered with unknown stress in the neuro-fuzzy techniques based real time PD recognition are also addressed.

• Structural Engineering and Mechanics
• /
• 제83권3호
• /
• pp.341-351
• /
• 2022

During bolted flange assembly, the contact stiffness of some areas of the joint surface may be low due to the elastic interaction. In order to improve the contact stiffness at the lowest position of bolted flange, the correlation model between the initial bolt pre-tightening force and the contact stiffness of bolted flange is established in this paper. According to the stress distribution model of a single bolt, an assumption of uniform local contact stiffness of bolted flange is made. Moreover, the joint surface is divided into the compressive stress region and the elastic interaction region. Based on the fractal contact theory, the relationship model of contact stiffness and contact force of the joint surface is proposed. Considering the elastic interaction coefficient method, the correlation model of the initial bolt pre-tightening force and the contact stiffness of bolted flange is established. This model can be employed to reverse determine the tightening strategy of the bolt group according to working conditions. As a result, this provides a new idea for the digital design of tightening strategy of bolt group for contact stiffness of bolted flange. The tightening strategy of the bolted flange is optimized by using the correlation model of initial bolt pre-tightening force and the contact stiffness of bolted flange. After optimization, the average contact stiffness of the joint surface increased by 5%, and the minimum contact stiffness increased by 6%.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 1999년도 추계학술대회-지능형 정보기술과 미래조직 Information Technology and Future Organization
• /
• pp.365-373
• /
• 1999

Recently, numerous studies have demonstrated that artificial intelligence such as neural networks can be an alternative methodology for classification problems to which traditional statistical methods have long been applied. In building neural network model, the selection of independent and dependent variables should be approached with great care and should be treated as a model construction process. Irrespective of the efficiency of a learning procedure in terms of convergence, generalization and stability, the ultimate performance of the estimator will depend on the relevance of the selected input variables and the quality of the data used. Approaches developed in statistical methods such as correlation analysis and stepwise selection method are often very useful. These methods, however, may not be the optimal ones for the development of neural network models. In this paper, we propose a genetic algorithms approach to find an optimal or near optimal input variables for neural network modeling. The proposed approach is demonstrated by applications to bankruptcy prediction modeling. Our experimental results show that this approach increases overall classification accuracy rate significantly.

• Journal of Mechanical Science and Technology
• /
• 제19권7호
• /
• pp.1393-1404
• /
• 2005

Fatigue crack growth and life have been estimated based on established empirical equations. In this paper, an alternative method using artificial neural network (ANN) -based model developed to predict fatigue damages simultaneously. To learn and generalize the ANN, fatigue crack growth rate and life data were built up using in-plane bending fatigue test results. Single fracture mechanical parameter or nondestructive parameter can't predict fatigue damage accurately but multiple fracture mechanical parameters or nondestructive parameters can. Existing fatigue damage modeling used this merit but limited real-time damage monitoring. Therefore, this study shows fatigue damage model using backpropagation neural networks on the basis of X -ray half breadth ratio B / $B_o$, fractal dimension $D_f$ and fracture mechanical parameters can estimate fatigue crack growth rate da/ dN and cycle ratio N / $N_f$ at the same time within engineering limit error ($5\%$).

• 동의생리병리학회지
• /
• 제22권5호
• /
• pp.1057-1061
• /
• 2008

For the purpose of investigating the reasonable logics contained in physiognomy of east and old western medicine. hypothetical researches based on hydromechanics theory were performed concerning facial types of form and pathologic features, especially 4 types of Dr. Jisan-Essence, Qi energy. Emotional Activity and Blood(EQAB). In order to infer the functional relation between facial type forming and EQAB factors, EQAB were supposed as fluid grounded on their continual flowing or periodical change and pressure effect from its congestion. and a premise that there's a linear corresponding relationship between the appearance of organ and its physical conditions of its inner vessels is formed too. Through this work, the unit fluid model(UFM) of Essence can be assumed as circle shape formed by the high viscosity and surface tension, the UFM model of Qi energy as quadrangular shape by the scattering features to outer four directions, and the UFM of emotional activity as inverted triangular shape by the flippant and uprising features, and the UFM of blood as ellipsoid triangle by the heavy and descending features in spite of circulation. The shapes made from each UFM are reproduced in the process of human development and manifest respective facial shape through the self-reproduction method like fractal theory in the last. Conclusively. it is said that the facial form analysis method like EQAB type theory can be the useful methodology to understand the human pathological and physiological features in view of hydromechanics.

• 지능정보연구
• /
• 제5권1호
• /
• pp.103-123
• /
• 1999

Input filtering as a preprocessing method is so much crucial to get good performance in time series forecasting. There are a few preprocessing methods (i.e. ARMA outputs as time domain filters, and Fourier transform or wavelet transform as time-frequency domain filters) for handling time series. Specially, the time-frequency domain filters describe the fractal structure of financial markets better than the time domain filters due to theoretically additional frequency information. Therefore, we, first of all, try to describe and analyze specially some issues on the effectiveness of different filtering methods from viewpoint of the performance of a neural network based forecasting. And then we discuss about neural network model architecture issues, for example, what type of neural network learning architecture is selected for our time series forecasting, and what input size should be applied to a model. In this study an input selection problem is limited to a size selection of the lagged input variables. To solve this problem, we simulate on analyzing and comparing a few neural networks having different model architecture and also use an embedding dimension measure as chaotic time series analysis or nonlinear dynamic analysis to reduce the dimensionality (i.e. the size of time delayed input variables) of the models. Throughout our study, experiments for integration methods of joint time-frequency analysis and neural network techniques are applied to a case study of daily Korean won / U. S dollar exchange returns and finally we suggest an integration framework for future research from our experimental results.

• 디지털융복합연구
• /
• 제11권8호
• /
• pp.205-212
• /
• 2013

본 논문에서는 한국의 전통 돌담을 생성하는 프랙탈을 이용한 절차적 모델링 알고리즘을 제시한다. 각 단계의 삼각형 메쉬를 세분하여 다음 단계의 메쉬를 얻는 것이 알고리즘의 주요 내용이다. 메쉬의 세분화 과정은 원하는 단계까지 재귀적으로 적용된다. 삼각형의 각 변에서 세 점을 무작위로 선택하여 네 작은 삼각형으로 나눈 후 각 정점이 법벡터 방향으로 조금씩 이동하면서 돌담의 울퉁불퉁한 모양을 만들어낸다. 매 단계마다 세분화한 메쉬에서 모양에 변화를 주지 않는 평평한 정점을 제거한다. 평평한 정도는 이산 곡률로 측정한다. 제거된 정점 주변의 점들로부터 새로운 삼각형들을 만들어 삼각형 메쉬를 완성한다.

• Geomechanics and Engineering
• /
• 제16권1호
• /
• pp.85-95
• /
• 2018

Water affects the mechanical properties of coal and stress wave propagation. To comprehensively investigate the effect of water content on the properties of coal, laboratory tests including X-Ray Diffraction (XRD) analysis, P-wave test, S-wave test, static and dynamic compression test with different water contents were conducted. The compressive strength, elastic modulus and failure strain and their mechanism of coal specimen under coupled static-dynamic load with the increased water content were observed. Meanwhile, energy transmission and dissipation characteristics of a stress wave in coal specimens with different water contents under dynamic load and its relation with the failure features, such as fragmentation and fractal dimension, of coal was analyzed. Furthermore, the dynamic interpretation of water infusion to prevent coal burst based on water infusion model of coal seam roadway was provided.

• Environmental Engineering Research
• /
• 제21권2호
• /
• pp.203-210
• /
• 2016

Hybrid materials were obtained modifying the bentonite (BC) and local clay (LC) using hexadecyltrimethylammonium bromide (HDTMA) or the clay were pillared with aluminum followed by modification with HDTMA. The materials were characterized by the SEM, FT-IR and XRD analytical tools. The batch reactor data implied that the uptake of $17{\beta}$-estradiol (E2) by the hybrid materials showed very high uptake at the neutral pH region. However, at higher and lower pH conditions, slightly less uptake of E2 was occurred. The uptake of E2 was insignificantly affected changing the sorptive concentration from 1.0 to 10.0 mg/L and the background electrolyte (NaCl) concentrations from 0.0001 to 0.1 mol/L. Moreover, the sorption of E2 by these hybrid materials was fairly efficient since within 30 mins of contact time, an apparent equilibrium between solid and solution was achieved, and the data was best fitted to the PSO (pseudo-second order) and FL-PSO (Fractal-like-pseudo second order) kinetic models compared to the PFO (pseudo-first order) model. The fixed-bed column results showed that relatively high breakthrough volume was obtained for the attenuation of E2 using these hybrid materials, and the loading capacity of E2 was estimated to be 75.984, 63.757, 58.965 and 49.746 mg/g for the solids BCH, BCAH, LCH and LCAH, respectively.

• 한국지능시스템학회논문지
• /
• 제8권2호
• /
• pp.82-95
• /
• 1998

본 논문에서는 구문독립 화지인식 시스템에서 가장 중요한 역할을 하는 분류기를 두 단계로 나누어, 먼저 짧은 구간들에 대해서 각각의 화자에 속하는 정도를 계산하고, 다음에 계산된 결과들을 가지고 주어진 음성구간전체에 대해 가장 가능성이 높은 화자를 선택하는 구조를 제안한다. 첫번째 부분은 학습에 의해 스스로 조기하는 RBFN을 이용하여 구현하고 두번째 부분에서는 MAXNET과 지연합의 조합으로 화자를 결정한다. 이렇게 함으로써 지연합의 개수가 증가함에 따라 인식률이 100%가 되는 것을 모의 실험을 통하여 확인한다. 또한 본 논문에서는 음성의 프랙탈적인 특징이 화자인식에 사용될 수 있는지를 검토한다. 화자인식은 동질의 집단에서 13명의 성인만자의 목소리를 이용하여 닫힌집합(closed-set)의 경우로 모의실험을 하였고, 기존의 특징으로는 선형예측계수(LPC) 와 PC-cepstrum을 사용하였다.