• Proceedings of the KSR Conference
• /
• 2008.06a
• /
• pp.841-848
• /
• 2008

This paper examined dynamic characteristics of high speed trains using a time varying frequency transform. Fourier transform based methods are frequently used for the calculation of the dynamic characteristics of trains in the frequency domain, but they cannot represent the time-varying characteristics. Therefore it is necessary to examine their characteristics using a time-varying frequency transform. For the examination, the non-stationary vibration of wheelset, bogie, and carbody are measured using accelerometers and stored in a data aquisition system. They are processed with localization of the data by modulating with a window function, and Fourier transform is taken to each localized data, called the short-time Fourier transform. From the processed results, time varying auto-spectral density, cross-spectral density, frequency response, and coherence functions have been calculated. From the analysis, it is confirmed that the time varying frequency transform is a useful method for analyzing the dynamic characteristics of high speed trains.

• The Journal of the Acoustical Society of Korea
• /
• v.35 no.2
• /
• pp.102-109
• /
• 2016

This paper proposes a noise suppression method through SNR (Singal-to Noise Ratio) estimation in the two microphone array environment of close spacing. The conventional method uses a noise suppression method for a gain function obtained through the SNR estimation based on coherence function from full band. However, this method cause performance decreased by the noise damage that affects all the feature vector component. So, we propose a noise suppression method that allocates a frequency domain signal into N constant multi frequency band and each frequency band gets a gain function through SNR estimation based on coherence function. Performance evaluation of the proposed method is shown by comparison with PESQ (Perceptual Evaluation of Speech Quality) value which is an objective quality evaluation method provided by the ITU-T (International Telecommunications Union Telecommunication).

• Journal of Korea Multimedia Society
• /
• v.7 no.12
• /
• pp.1639-1649
• /
• 2004

In this paper, a method for detection of forged signatures based on spectral analysis of directional gradient density function and a weighted fuzzy classifier is proposed. The well defined outline of an incoming signature image is extracted in a preprocessing stage which includes noise reduction, automatic thresholding, image restoration and erosion process. The directional gradient density function derived from extracted signature outline is highly related to the overall shape of signature image, and thus its frequency spectrum is used as a feature set. With this spectral feature set, having a property to be invariant in size, shift, and rotation, a weighted fuzzy classifier is evaluated for the verification of freehand and random forgeries. Experiments show that less than 5% averaged error rate can be achieved on a database of 500 signature samples.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2009.06a
• /
• pp.225-226
• /
• 2009

• Structural Engineering and Mechanics
• /
• v.39 no.1
• /
• pp.47-75
• /
• 2011

An accurate substructural synthesis method including random responses synthesis, frequency-response functions synthesis and mid-order modes synthesis is developed based on rigorous substructure description, dynamic condensation and coupling. An entire structure can firstly be divided into several substructures according to different functions, geometric and dynamic characteristics. Substructural displacements are expressed exactly by retained mid-order fixed-interfacial normal modes and residual constraint modes. Substructural interfacial degree-of-freedoms are eliminated by interfacial displacements compatibility and forces equilibrium between adjacent substructures. Then substructural mode vibration equations are coupled to form an exact-condensed synthesized structure equation, from which structural mid-order modes are calculated accurately. Furthermore, substructural frequency-response function equations are coupled to yield an exact-condensed synthesized structure vibration equation in frequency domain, from which the generalized structural frequency-response functions are obtained. Substructural frequency-response functions are calculated separately by using the generalized frequency-response functions, which can be assembled into an entire-structural frequency-response function matrix. Substructural power spectral density functions are expressed by the exact-synthesized substructural frequency-response functions, and substructural random responses such as correlation functions and mean-square responses can be calculated separately. The accuracy and capacity of the proposed substructure synthesis method is verified by numerical examples.

• Journal of Power Electronics
• /
• v.16 no.3
• /
• pp.849-860
• /
• 2016

The high-switching-frequency operation of power converters can achieve high power density through size reduction of passive components, such as capacitors, inductors, and transformers. However, a small-output capacitor that has small capacitance and low effective series resistance changes the small-signal model of the converter power stage. Such a capacitor can make the converter unstable by increasing the crossover frequency in the transfer function of the small-signal model. In this paper, the design and implementation of a high-frequency LLC resonant converter are presented to verify the power density enhancement achieved by decreasing the size of passive components. The effect of small output capacitance is analyzed for stability by using a proper small-signal model of the LLC resonant converter. Finally, proper design methods of a feedback compensator are proposed to obtain a sufficient phase margin in the Bode plot of the loop gain of the converter for stable operation at 500 kHz switching frequency. A theoretical approach using MATLAB, a simulation approach using PSIM, and experimental results are presented to show the validity of the proposed analysis and design methods with 100 and 500 kHz prototype converters.

• Water Engineering Research
• /
• v.1 no.1
• /
• pp.49-61
• /
• 2000

Magnitude-frequency relationships are used in the design of dams, highway bridges, culverts, water supply systems, and flood control structures. In this paper, possible techniques for analyzing flood frequency at a site are presented. A currently used approach to flood frequency analysis is based on the concept of parametric statistical inference. In this analysis, the assumption is make that the distribution function describing flood data in known. However, such an assumption is not always justified. Even though many people have shown that the nonparametric method provides a better fit to the data than the parometric method and gives more reliable flood estimates. the noparpmetric method implies a small probability in extrapolation beyond the highest observed data in the sample. Therefore, a remedy is presented in this paper by introducing an estimator which mixes parametric and nonparametric density estimate.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.1A
• /
• pp.59-67
• /
• 2008

Acceleration time history used in the seismic analysis of nuclear power plant structures should envelop a target power spectral density (PSD) function in addition to the design response spectrum in order to have sufficient energy at each frequency for the purpose of ensuring adequate load. The safety for complex long-span highway bridges cannot be over-emphasize. An alternative method to improve the seismic capacity is to ensure the minimum PSD function of the applied seismic load. This study proposes a technical scheme to obtain the reference power spectral density function by using artificial earthquakes which are compatible with the highway bridge design spectrum.

• Proceedings of the KSME Conference
• /
• 2000.04b
• /
• pp.311-315
• /
• 2000

A numerical investigation is made of time-dependent buoyant convection in a square of a non-Boussinesq fluid. The density-temperature$({\rho}-T)$ relation is modeled by a quadratic function, with the maximum density ${\rho}_M$ at temperature $T_M$. The horizontal walls of the square are insulated, and a pulsating temperature $T_H=T_M+{\Delta}T'\;sin({\omega}{\tau})$ is imposed on the hot vertical sidewall. The temperature at the cold wall $T_c$ is constant. Extensive numerical solutions to the governing Navier-Stokes equations are portrayed. Resonance is identified by monitoring the amplitude of the mid-plane Nusselt number, $A(Nu^*)$. The primary resonance frequency is found by matching ${\omega}$ to the nondimensional basic mode $N_1$ of internal gravity oscillations. Due to the quadratic$({\rho}-T)$ relationship, the effective pulsation frequency for density, $2{\omega}$, is meaningful, which brings forth the secondary resonance frequency, i.e., $2{\omega}=N_1$

• Geotechnical Engineering
• /
• v.4 no.3
• /
• pp.5-18
• /
• 1988

This study aims at scrutinizing the relationships among the frequency distribution of voids the relative density and the shear strength for the three types of different granular materials. The main results summarized are as follows: (1) The frequency distribution of void ratio of glass beads can be approximated by the negative exponential distribution proposed by Shahinpoor (1981), while as the particle shape changes from spherical to angular the frequency of the denser voids decreases and the distributions are well presented by the beta distribution. (2) For all materials, the standard deviation of void ratio increases with decreasing both the relative density of the material and sphericity of particle. (3) It was found that shear strength of the material was a function of not only the mean void ratio and its standard deviation, but also of the shape of the probability density function of the distribution of void ratio. The more the frequency distribution of void inclines towards the denser voids, the higher the shear strength of the material and vice versa.