• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2003년도 봄 학술발표회 논문집
• /
• pp.357-365
• /
• 2003

The major problems with the conventional neural network, especially Back Propagation Neural Network, arise from the necessity of many training data for neural network learning and ambiguity in the relation of neural network structure to the convergence of solution. In this paper, the PNN is used as a pattern classifier to detect the damage of structure to avoid those drawbacks of the conventional neural network. In the PNN-based pattern classification problems, the probability density function for patterns is usually assumed by Gaussian distribution. But, in this paper, several probability density functions are investigated in order to select the most approriate one for structural damage assessment.

• 한국전기전자재료학회논문지
• /
• 제11권4호
• /
• pp.261-266
• /
• 1998

In developing the high density memory device, the etching of fine pattern is becoming increasingly important. Therefore, definition of ultra fine line and space pattern and minimization of damage and contamination are essential process. Also, the high density plasma in low operating pressure is necessary. The candidates of high density plasma sources are electron cyclotron resonance plasma, helicon wave plasma, helical resonator, and inductively coupled plasma. In this study, planar type magnetized inductively coupled plasma etcher has been built. The density and temperature of Ar plasma are measured as a function of rf power, flow rate, external magnetic field, and pressure. The oxide etch rate and selectivity to polysilicon are measured as the above mentioned conditions and self-bias voltage.

• 한국멀티미디어학회논문지
• /
• 제25권4호
• /
• pp.620-627
• /
• 2022

This study analyzed the shot pattern through the tempo of segmented shot duration and studied the relationship with the unique emotion of the story. The structure of the story was classified into 3 chapters, 17 sequences, 83 scenes, 287 beats, and 1636 shots. Shot density is a method of visualizing tension in visual storytelling, and since it is a result obtained by mathematically calculating the density of divided shots, it can be helpful in designing tension delivered to the audience. Nine shot density patterns were extracted. The ascending(+) type was classified as A, B, C, D, 4, the descending(-) type, E, F, G, H, 4, and the maintenance(/) type, I, 1 type. Based on the spatiality of the 17 stages of Campbell's heroic narrative and McGee's story structure, the narrative level of the tree structure was proposed, and the symbolic meaning of the shot rhythm in the practical aspect of the story function was proposed to present a systematic methodology in the direction of production.

• The Korean Journal of Ecology
• /
• 제13권2호
• /
• pp.83-92
• /
• 1990

Two-dimensional analysis provides a comprehensive description of the structure, scales of pattern and directional components in a spatial data set. In spectral analysisi, four functions are illustrated,; the autocorrelation, the periodogram, the R-spectrum and the $\theta$ -spectrum. The R-spectrum and $\theta$ -spectrum function respectively summarize the periodogram in term of scale of pattern and directional components. Sampling is measured in the Naejang National Park area where the Daphniphyllum trees grow. 320 contiguous (15$\times$15)m plots are located along the transect and density of all trees over DBH 3 cm recorded respectively. 12 species of vascular plant are recorded in this survey area. The trend surface of density of all plant are estimated using polynomial regression and are exhibited in 3-dimensional graph and density contour map. Transformation to the corresponding polar spectrum from the periodogram emphasized the directional components and the scales to pattern. R-spectrum corresponding to the scale of pattern of periodogram showed a large peak 15.47 in the interval 9$\theta$-spectrum corresponding to directional components have two peaks 8.28 and 11.05 in the interval $35^{\circ}\theta <45^{\circ}and 125^{\circ}\theta< <135^{\circ}, respectively. Programs to compute all the analyses described in this study was obtained from Dr. Ranshow and was translated to BASIC by the author.

• Tribology and Lubricants
• /
• 제26권6호
• /
• pp.311-316
• /
• 2010

Surface texturing of micro dimple or pore-shaped pattern was prepared using a fiber laser system. Surface texturing was designed to have a square pattern with a particular pitch distance for each corresponding density of 5, 10, 20, and 30%. Thermal damages such as bulges and burrs formed during laser irradiation were observed around the dimples. Thermal damages were later removed by lapping using alumina particles of $0.3{\mu}m$ in diameter. Oscillating friction tests were performed against heat-treated high speed steels under lubricated condition. The lubricant used was SAE 5W-30 automotive engine oil. Normal contact pressure and oscillating frequency was 0.28 MPa and 20 Hz, respectively. The tests were continued for 20 minutes, and friction plots were recorded for examination. Results revealed that the coefficient of friction was lowered regardless of texturing density. Moreover, the lowest coefficient of friction was obtained for 10% density texturing. It is attributed to increased lubricity due to the introduction of surface texturing. In addition, it is concluded that the optimum texturing density and pattern must be found for the best lubricity and low friction.

• Journal of Mechanical Science and Technology
• /
• 제17권7호
• /
• pp.966-977
• /
• 2003

This paper investigates how intentional mistuning of bladed disks reduces their sensitivity to unintentional random mistuning. The class of intentionally mistuned disks considered here is limited, for cost reasons, to arrangements of two types of blades (A and B, say). A two-step procedure is then described to optimize the arrangement of these blades around the disk to reduce the effects of unintentional random mistuning. First, a pure optimization effort is undertaken to obtain the pattern (s) of the A and B blades that yields small/the smallest value of the largest amplitude of response to a given excitation in the absence of unintentional random mistuning using Genetic Algorithm. Then, in the second step, a qualitative/quantitative estimate of the sensitivity for the optimized intentionally mistuned bladed disks with respect to unintentional random mistuning is performed by analyzing their amplification factor, probability density function and passband/stopband structures. Examples of application with simple bladed disk models demonstrate the significant benefits of using this class of intentionally mistuned disks.

• Journal of Advanced Marine Engineering and Technology
• /
• 제15권3호
• /
• pp.45-56
• /
• 1991

This paper describes characteristics of the flow pattern of the flow past a rectangular cylinder experimentally investigated. The width-to-length ratio of the section varried from 2 to4. For the statistical treatment, autocorrelation coefficient, probability density function and power spectral density function are obtained by the digital processing technic through on-line system with a hot wire anemometer. As a results, it was found that strong periodic coherent eddies structure is sustained to about 20H downstream from the cylinder. And nearer the cylinder in the wake, the number of turbulent eddies of a large scale coherent structure are comparatively much more dominant than that of a small scale one. By the analysis of power spectrum, It was cleared that there exists a certain range of the width to length ratio between 2.5 and 3 of which the flow pattern changes abruptly with a sudden discontinuity in Strouhal number.

• 대한조선학회논문집
• /
• 제52권1호
• /
• pp.88-95
• /
• 2015

For the frequency-domain spectral fatigue analysis, the probability density function of stress range needs to be estimated based on the stress spectrum only, which is a frequency domain representation of the response. The probability distribution of the stress range of the narrow-band spectrum is known to follow the Rayleigh distribution, however the PDF of wide-band spectrum is difficult to define with clarity due to the complicated fluctuation pattern of spectrum. In this paper, efforts have been made to figure out the links between the probability density function of stress range to the structural response of wide-band Gaussian random process. An artificial neural network scheme, known as one of the most powerful system identification methods, was used to identify the multivariate functional relationship between the idealized wide-band spectrums and resulting probability density functions. To achieve this, the spectrums were idealized as a superposition of two triangles with arbitrary location, height and width, targeting to comprise wide-band spectrum, and the probability density functions were represented by the linear combination of equally spaced Gaussian basis functions. To train the network under supervision, varieties of different wide-band spectrums were assumed and the converged probability density function of the stress range was derived using the rainflow counting method and all these data sets were fed into the three layer perceptron model. This nonlinear least square problem was solved using Levenberg-Marquardt algorithm with regularization term included. It was proven that the network trained using the given data set could reproduce the probability density function of arbitrary wide-band spectrum of two triangles with great success.

• 노동경제논집
• /
• 제36권3호
• /
• pp.29-64
• /
• 2013

본 논문은 2000년대 임금밀도함수의 변화를 추정하고, 근로자 및 일자리 구성의 변화 등을 이용하여 구성 분해한 결과를 제시한다. 2000년대 변화의 특징은 임금 중위층의 위축, 상위층의 대폭 확대, 하위층의 소폭 확대이며, 이 가운데 1/3 정도를 근로자 구성 변화가 설명하지만, 일자리 구성 변화는 상대적으로 설명력이 약한 것으로 나타났다. 따라서 근로자 구성에 의해 설명되지 못하는 나머지 변화들은 일자리 유형 내에서 고기능 수요의 확대에 공급이 충분히 대응하지 못함을 반영하는 것으로 판단된다. 이러한 양상은 제조업, 대기업, 노조부문, 생산직 등에서 두드러지는데, 즉 이 부문들에서의 수요 변화가 전반적인 임금분포의 변화를 주도하고 있다고 판단된다.

• The Journal of the Acoustical Society of Korea
• /
• 제17권4E호
• /
• pp.3-10
• /
• 1998

This paper concerns the dynamical behavior, in probabilistic sense, of a feedforward neural network performing auto association for novelty. Networks of retinotopic topology having a one-to-one correspondence between and output units can be readily trained using back-propagation algorithm, to perform autoassociative mappings. A novelty filter is obtained by subtracting the network output from the input vector. Then the presentation of a "familiar" pattern tends to evoke a null response ; but any anomalous component is enhanced. Such a behavior exhibits a promising feature for enhancement of weak signals in additive noise. As an analysis of the novelty filtering, this paper shows that the probability density function of the weigh converges to Gaussian when the input time series is statistically characterized by nonsymmetrical probability density functions. After output units are locally linearized, the recursive relation for updating the weight of the neural network is converted into a first-order random differential equation. Based on this equation it is shown that the probability density function of the weight satisfies the Fokker-Planck equation. By solving the Fokker-Planck equation, it is found that the weight is Gaussian distributed with time dependent mean and variance.