• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.2B
• /
• pp.131-137
• /
• 2006

We have investigated the properties of the Singular Spectrum Analysis (SSA) coupled with the Linear Recurrent Formula which made it possible to complement the parametric time series model. The SSA has been applied to extract the underlying properties of the principal component of hydrologic time series, which can often be identified as trends, seasonalities and other oscillatory series, or noise components. Generally, the prediction by the SSA method can be applied to hydrologic time series governed (may be approximately) by the linear recurrent formulae. This study has examined the forecasting ability of the SSA-LRF model. These methods were applied to monthly discharge and water surface level data. These models indicated that two of the time series have good abilities of forecasting, particularly showing promising results during the period of one year. Thus, the method presented in this study suggests a competitive methodology for the forecast of hydrologic time series.

• Structural Engineering and Mechanics
• /
• v.39 no.6
• /
• pp.877-893
• /
• 2011

Response spectra of earthquake ground motions are important in the earthquake-resistant design and reliability analysis of structures. The formulation of the response spectrum in the frequency domain efficiently computes and evaluates the stochastic response spectrum. The frequency information of the excitation can be described using different functional forms. The shapes of the calculated response spectra of the excitation show strong magnitude and site dependency, but weak distance dependency. In this paper, to compare the effect of the earthquake ground motion variables, the contribution of these sources of variability to the response spectrum's uncertainty is calculated by using a stochastic analysis. The analytical results show that earthquake source factors and soil condition variables are the main sources of uncertainty in the response spectra, while path variables, such as distance, anelastic attenuation and upper crust attenuation, have relatively little effect. The presented formulation of dynamic structural response in frequency domain based only on the frequency information of the excitation can provide an important basis for the structural analysis in some location that lacks strong motion records.

• Journal of the Korean Society for Precision Engineering
• /
• v.23 no.1 s.178
• /
• pp.105-112
• /
• 2006

The nonlinear damping force model is made to identify the properties of the ER (electro-rheological) fluid suspension damper. The instrumentation is carried out to measure the damping force of the ER damper. The higher order spectral analysis method is used to investigate the nonlinear frequency coupling phenomena with the damping force signal according to the sinusoidal excitation of the damper. The distinctive higher order nonlinear characteristics are observed. The nonlinear damping force model, which has the higher order velocity terms, is proposed with the result of higher order spectrum analysis. The higher order terms coefficients, which vary according to the strength of the electric field, are calculated using the least square method.

• Journal of Biosystems Engineering
• /
• v.33 no.4
• /
• pp.269-275
• /
• 2008

This study was conducted to design an impedance sensor that can measure soil water content of soils. Partial least square regression (PLSR) was applied to soil impedance data preprocessed with a smoothing method. An optimal sub-spectrum size and wavelength range were determined by comparing the coefficient of determination ($R^2$) and root mean square error (RMSE) of the PLSR models obtained using soil impedance data. various PLS analysis. Based on the PLSR analysis, it would be concluded that the optimal spectrum measurement range was $32.0{\sim}50.0\;MHz$ with the optimal sub-spectrum size of about 18.5 MHz.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.3E
• /
• pp.89-94
• /
• 2006

For audio indexing and targeted search of specific audio or corresponding visual contents, the MPEG-7 standard has adopted a sound classification framework, in which dimension-reduced Audio Spectrum Projection (ASP) features are used to train continuous hidden Markov models (HMMs) for classification of various sounds. The MPEG-7 employs Principal Component Analysis (PCA) or Independent Component Analysis (ICA) for the dimensional reduction. Other well-established techniques include Non-negative Matrix Factorization (NMF), Linear Discriminant Analysis (LDA) and Discrete Cosine Transformation (DCT). In this paper we compare the performance of different dimensional reduction methods with Gaussian mixture models (GMMs) and HMMs in the classifying video sound clips.

• Journal of Industrial Technology
• /
• v.24 no.A
• /
• pp.195-204
• /
• 2004

To evaluate inelastic seismic responses of multi-span bridge structures, the capacity spectrum method(CSM) incorporating the equivalent single-degree-of freedom(ESDOF) method is presented. Application of the CSM incorporating the ESDOF method is illustrated by example analysis for symmetric and asymmetric bridge structures. To investigate an accuracy of the CSM, the maximum displacements estimated by the CSM are compared to those by inelastic time history analysis for several artificial earthquakes. The results show that the CSM provided conservative estimates of the maximum displacements for the symmetric and asymmetric bridge structures, and the trend of conservative estimates of the asymmetric bridge structures was significantly larger than that of the symmetric bridge structure.

• Proceedings of the KIEE Conference
• /
• 2005.10b
• /
• pp.29-31
• /
• 2005

Detection methods for atrial tachycardia and fibrillation on the time axis have the advantages of light operational load and are easy to apply to various applications. Despite these advantages, arrhythmia detection algorithm on the time axis cannot stand much noise such as motion artifacts, moreover the peak detection algorithm has high complexity. In this paper, we use a spectrum analysis method for the detection of atrial tachycardia and fibrillation. By applying spectrum analysis and digital filtering on obtained electrogram signals, we can diagnose heart arrhythmia without using peak detection algorithm.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2003.04a
• /
• pp.27-34
• /
• 2003

Capacity Spectrum Method (CSM) defined in ATC-40 or FEMA-273 is a most widely used static inelastic analysis method to evaluate the performance level of the existing structures. In CSM, however, uncertainties and errors exist when lateral forces such as earthquake and wind loads are analyzed into equivalent static loads. This paper examines the accuracy of CSM for different structural parameters, such as natural frequency, yield strength and hardening ratio, and various soil conditions by comparing the estimated values to exact solutions obtained by time history analysis. Results indicate that the accuracy of CSM, in general, is influenced mostly by hardening ratio.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 2005.03a
• /
• pp.396-404
• /
• 2005

Nonlinear responses of structures may be obtained through three different methods. They are time-history analysis techniques, response spectrum method, and R-factor method. The nonlinear response spectrum method is frequently used in the practice, because the time history analysis method is time-consuming and complicated. There are two different approaches in obtaining the nonlinear response spectrum, which results in "constant displacement ductility spectra" and "constant damage spectra", respectively. The nonlinear response spectra of the various time-histories had been computed and the results were comparatively evaluated in this study. The study results showed that the existing constant displacement ductility spectra can induce unconservative design especially for the structures on soft soil base. This unconservatism can be removed by using the newly proposed constant damage spectra.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.17 no.9
• /
• pp.1013-1021
• /
• 2004

This paper presents the results of spectrum analysis of leakage current waveforms on contaminated polymer insulators under fog conditions. This is a basic study on development of a monitoring for aging degree and surface pollution state using distortion rate(%) of 3rd and 5th harmonics according fundamental frequency[60Hz] under environment conditions. Arcing on the surface was always associated with distortion in the leakage current ; hence the low frequency components of the leakage current can be used as a means to evaluate the surface degradation. The results indicate that the distortion rate by spectrum analysis is available for the monitoring of contamination condition of EPDM Polymer insulators.