• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.34 no.1
• /
• pp.35-42
• /
• 2021

This paper presents an on-line finite element model updating method for in-service structures using measured data. Conventional updating methods, which are based on numerical optimization, are not efficient for on-line updating because they generally require repeated eigenvalue analyses until convergence criteria are met. The proposed method enables fully automated on-line finite element model updating, almost simultaneously with vibration measurement, without any user intervention or off-line procedures. The automated covariance-driven stochastic subspace identification (Cov-SSI) method is utilized to identify modal frequencies and vectors, and the identified modal data is fed to the neural network of the inverse eigenvalue function to produce the updated finite element model parameters. Numerical examples for a wind excited 20-story building structure shows that the proposed method can update the series of finite element model parameters automatically. It is also shown that sudden changes in the structural parameters can be detected and traced successfully.

• Smart Structures and Systems
• /
• v.29 no.3
• /
• pp.499-512
• /
• 2022

This work proposes a novel contactless vibration damping and thermal isolation tripod platform based on Superconducting Magnetic Levitation (SML). This prototype is suitable for cryogenic environments, where classical passive, semi active and active vibration isolation techniques may present tribological problems due to the low temperatures and/or cannot guarantee an enough thermal isolation. The levitating platform consists of a Superconducting Magnetic Levitation (SML) with inherent passive static stabilization. In addition, the use of Operational Modal Analysis (OMA) technique is proposed to characterize the transmissibility function from the baseplate to the platform. The OMA is based on the Stochastic Subspace Identification (SSI) by using the Expectation Maximization (EM) algorithm. This paper contributes to the use of SSI-EM for SML applications by proposing a step-by-step experimental methodology to process the measured data, which are obtained with different unknown excitations: ambient excitation and impulse excitation. Thus, the performance of SSI-EM for SML applications can be improved, providing a good estimation of the natural frequency and damping ratio without any controlled excitation, which is the main obstacle to use an experimental modal analysis in cryogenic environments. The dynamic response of the 510 g levitating platform has been characterized by means of OMA in a cryogenic, 77 K, and high vacuum, 1E-5 mbar, environment. The measured vertical and radial stiffness are 9872.4 N/m and 21329 N/m, respectively, whilst the measured vertical and radial damping values are 0.5278 Nm/s and 0.8938 Nm/s. The first natural frequency in vertical direction has been identified to be 27.39 Hz, whilst a value of 40.26 Hz was identified for the radial direction. The determined damping values for both modes are 0.46% and 0.53%, respectively.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.3
• /
• pp.326-334
• /
• 2022

In this paper, we propose speech enhancement algorithm as a pre-processing for robust speech recognition in noisy environments. Auxiliary-function-based Independent Vector Analysis (AuxIVA) is performed with weighted covariance matrix using time-varying variances with scaling factor from target masks representing time-frequency contributions of target speech. The mask estimates can be obtained using Neural Network (NN) pre-trained for speech extraction or diffuseness using Coherence-to-Diffuse power Ratio (CDR) to find the direct sounds component of a target speech. In addition, outputs for omni-directional noise are closely chained by sharing the time-varying variances similarly to independent subspace analysis or IVA. The speech extraction method based on AuxIVA is also performed in Independent Low-Rank Matrix Analysis (ILRMA) framework by extending the Non-negative Matrix Factorization (NMF) for noise outputs to Non-negative Tensor Factorization (NTF) to maintain the inter-channel dependency in noise output channels. Experimental results on the CHiME-4 datasets demonstrate the effectiveness of the presented algorithms.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.5
• /
• pp.401-408
• /
• 2013

Dual-layer differential codebook using 8-PSK (phase shift keying) constellation as its codeword elements, is proposed for Long term evolution (LTE) and/or LTE-Advanced systems. Due to the temporal correlation of the adjacent channel matrices, the consecutive precoding matrices are likely to be similar. This approach quantize only the differential information of the channel instead of the whole channel subspace, which virtually increase the codebook size to realize more accurate quantization of the channel. Especially, the proposed codebook has the same properties of LTE release-8 codebook that is, constant modulus, complexity reduction, and nested property. The mobile station can be designed by using less expensive non-linear amplifier utilizing constant modulus property. Computer simulations show that the capacity of the proposed dual-layer codebook performs almost 1.2dB better than those of any other non-differential codebooks with the same amount of feedback information.

• Structural Engineering and Mechanics
• /
• v.72 no.6
• /
• pp.689-712
• /
• 2019

Time-series models like AR-ARX and ARMAX, provide a robust way to capture the dynamic properties of structures, and their residuals can be effectively used as features for damage detection. Even though several research papers discuss the implementation of AR-ARX and ARMAX models for damage diagnosis, they are basically been exploited so far for detecting the time instant of damage and also the spatial location of the damage. However, the inverse problem associated with damage quantification i.e. extent of damage using time series models is not been reported in the literature. In this paper, an approach to detect the extent of damage by combining the ARMAX model by formulating the inverse problem as a multi-constrained optimization problem and solving using a newly developed hybrid adaptive differential search with dynamic interaction is presented. The proposed variant of the differential search technique employs small multiple populations which perform the search independently and exchange the information with the dynamic neighborhood. The adaptive features and local search ability features are built into the algorithm in order to improve the convergence characteristics and also the overall performance of the technique. The multi-constrained optimization formulations of the inverse problem, associated with damage quantification using time series models, attempted here for the first time, can considerably improve the robustness of the search process. Numerical simulation studies have been carried out by considering three numerical examples to demonstrate the effectiveness of the proposed technique in robustly identifying the extent of the damage. Issues related to modeling errors and also measurement noise are also addressed in this paper.

• Journal of KIISE:Software and Applications
• /
• v.34 no.7
• /
• pp.681-687
• /
• 2007

Face recognition technique is very popular for a personal information security and user identification in recent years. However, the face recognition system is very hard to be implemented due to the difficulty where change in illumination, pose and facial expression. In this paper, we consider that an illumination change causing the variety of face appearance, virtual image data is generated and added to the D-LDA which was selected as the most suitable feature extractor. A less sensitive recognition system in illumination is represented in this paper. This way that consider nature of several illumination directions generate the virtual training image data that considered an illumination effect of the directions and the change of illumination density. As result of experiences, D-LDA has a less sensitive property in an illumination through ORL, Yale University and Pohang University face database.

• Smart Structures and Systems
• /
• v.6 no.5_6
• /
• pp.461-480
• /
• 2010

This paper analyses the data collected from the $2^{nd}$ Jindo Bridge, a cable-stayed bridge in Korea that is a structural health monitoring (SHM) international test bed for advanced wireless smart sensors network (WSSN) technology. The SHM system consists of a total of 70 wireless smart sensor nodes deployed underneath of the deck, on the pylons, and on the cables to capture the vibration of the bridge excited by traffic and environmental loadings. Analysis of the data is performed in both the time and frequency domains. Modal properties of the bridge are identified using the frequency domain decomposition and the stochastic subspace identification methods based on the output-only measurements, and the results are compared with those obtained from a detailed finite element model. Tension forces for the 10 instrumented stay cables are also estimated from the ambient acceleration data and compared both with those from the initial design and with those obtained during two previous regular inspections. The results of the data analyses demonstrate that the WSSN-based SHM system performs effectively for this cable-stayed bridge, giving direct access to the physical status of the bridge.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.17 no.3
• /
• pp.27-33
• /
• 2007

Recently, a new hard problem on a two dimensional vector space called vector decomposition problem (VDP) was proposed by M. Yoshida et al. and proved that it is at least as hard as the computational Diffe-Hellman problem (CDHP) on a one dimensional subspace under certain conditions. However, in this paper we present the VDP relative to a specific basis can be solved in polynomial time although the conditions proposed by M. Yoshida on the vector space are satisfied. We also suggest strong instances based on a certain type basis which make the VDP difficult for any random vector relative to the basis. Therefore, we need to choose the basis carefully so that the VDP can serve as the underlying intractable problem in the cryptographic protocols.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.4A
• /
• pp.605-609
• /
• 2006

In this study, two kinds of smart accelerometers are investigated for the application of smart sensors to the structural health monitoring of infrastructures. Smart optical Fiber Bragg Grating (FBG) type and Micro-Electo-Mechanical System (MEMS) type accelerometers are selected for this study and the high sensitive ICP type accelerometer is used for the reference sensor. Small size shaking table tests were performed with 3-story shear building model using random input ground motions. The output only modal identification was carried out using stochastic subspace identification and the performances of sensors are compared in modal domain indirectly. The modal sensitivity method was applied to update the story stiffness of numerical model and the updated results were verified using the additional experiments for the same structure with additional mass.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1244-1244
• /
• 2001

In linear discriminant analysis there are two important properties concerning the effectiveness of discriminant function modeling. The first is the separability of the discriminant function for different classes. The separability reaches its optimum by maximizing the ratio of between-class to within-class variance. The second is the stability of the discriminant function against noises present in the measurement variables. One can optimize the stability by exploring the discriminant variates in a principal variation subspace, i. e., the directions that account for a majority of the total variation of the data. An unstable discriminant function will exhibit inflated variance in the prediction of future unclassified objects, exposed to a significantly increased risk of erroneous prediction. Therefore, an ideal discriminant function should not only separate different classes with a minimum misclassification rate for the training set, but also possess a good stability such that the prediction variance for unclassified objects can be as small as possible. In other words, an optimal classifier should find a balance between the separability and the stability. This is of special significance for multivariate spectroscopy-based classification where multicollinearity always leads to discriminant directions located in low-spread subspaces. A new regularized discriminant analysis technique, the principal discriminant variate (PDV) method, has been developed for handling effectively multicollinear data commonly encountered in multivariate spectroscopy-based classification. The motivation behind this method is to seek a sequence of discriminant directions that not only optimize the separability between different classes, but also account for a maximized variation present in the data. Three different formulations for the PDV methods are suggested, and an effective computing procedure is proposed for a PDV method. Near-infrared (NIR) spectra of blood plasma samples from mastitic and healthy cows have been used to evaluate the behavior of the PDV method in comparison with principal component analysis (PCA), discriminant partial least squares (DPLS), soft independent modeling of class analogies (SIMCA) and Fisher linear discriminant analysis (FLDA). Results obtained demonstrate that the PDV method exhibits improved stability in prediction without significant loss of separability. The NIR spectra of blood plasma samples from mastitic and healthy cows are clearly discriminated between by the PDV method. Moreover, the proposed method provides superior performance to PCA, DPLS, SIMCA and FLDA, indicating that PDV is a promising tool in discriminant analysis of spectra-characterized samples with only small compositional difference, thereby providing a useful means for spectroscopy-based clinic applications.