• The Journal of the Acoustical Society of Korea
• /
• v.24 no.4
• /
• pp.210-216
• /
• 2005

We Propose a new feature normalization scheme based on eigenspace for achieving robust speech recognition. In general, mean and variance normalization (MVN) is Performed in cepstral domain. However, another MVN approach using eigenspace was recently introduced. in that the eigenspace normalization Procedure Performs normalization in a single eigenspace. This Procedure consists of linear PCA matrix feature transformation followed by mean and variance normalization of the transformed cepstral feature. In this method. 39 dimensional feature distribution is represented using only a single eigenspace. However it is observed to be insufficient to represent all data distribution using only a sin91e eigenvector. For more specific representation. we apply unique na independent eigenspaces to cepstra, delta and delta-delta cepstra respectively in this Paper. We also normalize training data in eigenspace and get the model from the normalized training data. Finally. a feature space rotation procedure is introduced to reduce the mismatch of training and test data distribution in noisy condition. As a result, we obtained a substantial recognition improvement over the basic eigenspace normalization.

• Structural Engineering and Mechanics
• /
• v.90 no.6
• /
• pp.551-568
• /
• 2024

The present study deals with static and dynamic behaviors including forced vibrations of an elastic rectangular nano plate on the two-parameter foundation. Firstly, the rectangular plate is assumed to be subjected to uniformly distributed and eccentrically applied concentrated loads. The governing equations of the problem are derived by considering the dynamic response of the plate, employing a series of the Chebyshev polynomials for the displacement function and applying the Galerkin method. Then, effects of the non-essential boundary conditions of the plate, i.e., the boundary conditions related to the shearing forces, the bending moments and the corner forces, are included in the governing equation of motion to compensate for the non-satisfied boundary conditions and increase the accuracy of the Galerkin method. The approximate numerical solution is accomplished using an iterative process due to the non-linearity of the unilateral property of the two-parameter foundation. The plate under static concentrated load is investigated in detail numerically by considering a wide range of parameters of the plate and the foundation stiffnesses. Numerical treatment of the problem in the time domain is carried out by assuming a stepwise variation of the concentrated load and the linear acceleration procedure is employed in the solution of the system of governing differential equations derived from the equation of motion. Time variations of the contact region and those of the displacements of the plate are presented in the figures for various numbers of the two-parameter of the foundation, as well as the classical and nano parameters of the plate particularly focusing on the non-linearity of the problem due to the plate lift-off from the unilateral foundation. The effects of classical and nonlocal parameters and loading are investigated in detail. Definition of the separation between the plate and the two-parameter foundation is presented and applied to the given problem. The effect of the lift-off on the static and dynamic behavior of the rectangular plate is studied in detail by considering various loading conditions. The numerical study shows that the effect of nonlocal parameters on the behavior of the plate becomes significant, when nonlinearity becomes more profound, due to the lift-off of the plate. It is seen that the size effects are significant in static and dynamic analysis of nano-scaled rectangular plates and need to be included in the mechanical analyses. Furthermore, the corner displacement of the plate is affected more significantly from the lift-off, whereas it is less marked in the time variation of the middle displacement of the plate. Several numerical examples are presented to examine the sensibility of various parameters associated with nonlocal parameters of the plate and foundation. Both stiffening and softening nonlocal parameters behavior of the plate are identified in the numerical solutions which show that increasing the foundation stiffness decreases the extent of the contact region, whereas the stiffness of the shear layer increases the contact region and reduces the foundation settlement considerably.

• Journal of Ocean Engineering and Technology
• /
• v.25 no.2
• /
• pp.134-144
• /
• 2011

In this paper, the global study trends for material behaviors are investigated regarding the static and dynamic hardenings and final fractures of marine structural steels. In particular, after reviewing all of the papers published at the 4th and 5th ICCGS (International Conference on Collision and Grounding of Ship), the used hardening and fracture properties are summarized, explicitly presenting the material properties. Although some studies have attempted to employ new plasticity and fracture models, it is obvious that most still employed an ideal hardening rule such as perfect plastic or linear hardening and a simple shear fracture criterion with an assumed value of failure strain. HSE (2001) presented pioneering study results regarding the temperature dependency of material strain hardening at various levels of temperature, but did not show strain rate hardening at intermediate or high strain rate ranges. Nemat-Nasser and Guo (2003) carried out fully coupled tests for DH-36 steel: strain hardening, strain rate hardening, and temperature hardening and softening at multiple steps of strain rates and temperatures. The main goal of this paper is to provide the theoretical background for strain and strain rate hardening. In addition, it presents the procedure and methodology needed to derive the material constants for the static hardening constitutive equations of Ludwik, Hollomon, Swift, and Ramberg-Osgood and for the dynamic hardening constitutive equations of power from Cowper-Symonds and Johnson-Cook.

• Advances in nano research
• /
• v.16 no.3
• /
• pp.289-301
• /
• 2024

In this work, an analytical model employing a new higher-order shear deformation beam theory is utilized to investigate the bending behavior of axially randomly oriented functionally graded carbon nanotubes reinforced composite nanobeams. A modified continuum nonlocal strain gradient theory is employed to incorporate both microstructural effects and geometric nano-scale length scales. The extended rule of mixture, along with molecular dynamics simulations, is used to assess the equivalent mechanical properties of functionally graded carbon nanotubes reinforced composite (FG-CNTRC) beams. Carbon nanotube reinforcements are randomly distributed axially along the length of the beam. The equilibrium equations, accompanied by nonclassical boundary conditions, are formulated, and Navier's procedure is used to solve the resulting differential equation, yielding the response of the nanobeam under various mechanical loadings, including uniform, linear, and sinusoidal loads. Numerical analysis is conducted to examine the influence of inhomogeneity parameters, geometric parameters, types of loading, as well as nonlocal and length scale parameters on the deflections and stresses of axially functionally graded carbon nanotubes reinforced composite (AFG CNTRC) nanobeams. The results indicate that, in contrast to the nonlocal parameter, the beam stiffness is increased by both the CNTs volume fraction and the length-scale parameter. The presented model is applicable for designing and analyzing microelectromechanical systems (MEMS) and nanoelectromechanical systems (NEMS) constructed from carbon nanotubes reinforced composite nanobeams.

• Journal of the Korea Concrete Institute
• /
• v.11 no.5
• /
• pp.89-98
• /
• 1999

This paper evaluates nonlinear response characteristics of precast concrete frame buildings. where plastics hinging occurs in the precast connection. Designs were developed for buildings of 5, 10 and 15 stories in hight for moderate seismic risk regions of the U. S. The responses of the buildings were analyzed using DRAIN-2DX and following Nonlinear static analysis procedure of ATC 19. The main variables of the analyses were the strength and stiffness of the connection. Also, for the analysis, the bi-linear response model, developed and inserted into the DRAIN-2DX program by Shan Shi and D. Fouch, was used. With the results of analysis, the deformation demands of the connection of precast concrete frame buildings are proposed by using equal-dissipated energy capacity. It was shown that the strength of the buildings as well as their displacement capacities decreased with the decrease of either the strength or stiffness in the connections. Therefore such changes also require reductions in the response modification factors for such buildings. However, if the precast concrete frame building has plastic hinging in the connection, and has a more ductile connection than the monolithic frame building, then no reduction in R may be necessary. The deformation demand required of the connection to achieve that condition is evaluated and a simple relation is suggested in the paper.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.3
• /
• pp.632-640
• /
• 2007

The low force estimation method of skeletal muscle was proposed by using ICA(independent component analysis) and neuro-transmission model. An EMG decomposition is the procedure by which the signal is classified into its constituent MUAP(motor unit action potential). The force index of electromyography was due to the generation of MUAP. To estimate low force, current analysis technique, such as RMS(root mean square) and MAV(mean absolute value), have not been shown to provide direct measures of the number and timing of motoneurons firing or their firing frequencies, but are used due to lack of other options. In this paper, the method based on ICA and chemical signal transmission mechanism from neuron to muscle was proposed. The force generation model consists of two linear, first-order low pass filters separated by a static non-linearity. The model takes a modulated IPI(inter pulse interval) as input and produces isometric force as output. Both the step and random train were applied to the neuro-transmission model. As a results, the ICA has shown remarkable enhancement by finding a hidden MAUP from the original superimposed EMG signal and estimating accurate IPI. And the proposed estimation technique shows good agreements with the low force measured comparing with RMS and MAV method to the input patterns.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.22 no.2
• /
• pp.27-33
• /
• 2014

There are several methods to improve the flight efficiency of HALE(High Altitude Long Endurance) UAV(Unmaned Aerial Vehicle). Airframe structural point of view, weight reduction of the airframe structure is the most important method to improve the flight efficiency. In order to reduce the weight of airframe structures, new concepts which are different from traditional airframe structure design such as the mylar wing skin should be introduced. The spar is the most important component in a mylar skin wing structure, so the spar weight reduction is the key point for reduction of the wing structural weight. In this study, design trade-off study for the front spar of the HALE UAV wing is conducted in order to reduce the weight. Design and analysis procedure of high aspect ratio wing spar are introduced. Several front spar structures are designed and trade-off study regarding the weight and strength for the each spar are performed. Spar design configurations are verified by the static strength test. Finally, optimal front spar design is decided and applied to the HALE UAV wing design.

• Structural Engineering and Mechanics
• /
• v.38 no.1
• /
• pp.27-37
• /
• 2011

This paper presents the results of a study on the capability of nonlinear quasi-static finite element modelling in simulating the hysteretic behaviour of CFRP and GFRP-retrofitted RC exterior beam-column joints under cyclic loads. Four specimens including two plain and two CFRP/GFRP-strengthened beam-column joints tested by Mahini and Ronagh (2004) and other researchers are modelled using ANSYS. Concrete in compression is defined by the modified Hognestad model and anisotropic multi-linear model is employed for modelling the stress-strain relations in reinforcing bars while anisotropic plasticity is considered for the FRP composite. Both concrete and FRP are modelled using solid elements whereas space link elements are used for steel bars considering a perfect bond between materials. A step by step load increment procedure to simulate the cyclic loading regime employed in the testing. An automatically reforming stiffness matrix strategy is used in order to simulate the actual seismic performance of the RC concrete after cracking, steel yielding and concrete crushing during the push and pull loading cycles. The results show that the hysteretic simulation for all specimens is satisfactory and therefore suggest that the numerical model can be used as an inexpensive tool to design of FRP-strengthened RC beam-column joints under cyclic loads.

• Structural Engineering and Mechanics
• /
• v.52 no.1
• /
• pp.149-172
• /
• 2014

Four algorithms for damage detection of trusses are presented in this paper. These approaches can detect damage by using both complete and incomplete measurements. The suggested methods are based on the minimization of the difference between the measured and analytical static responses of structures. A non-linear constrained optimization problem is established to estimate the severity and location of damage. To reach the responses, the successive quadratic method is used. Based on the objective function, the stiffness matrix of the truss should be estimated and inverted in the optimization procedure. The differences of the proposed techniques are rooted in the strategy utilized for inverting the stiffness matrix of the damaged structure. Additionally, for separating the probable damaged members, a new formulation is proposed. This scheme is employed prior to the outset of the optimization process. Furthermore, a new tactic is presented to select the appropriate load pattern. To investigate the robustness and efficiency of the authors' method, several numerical tests are performed. Moreover, Monte Carlo simulation is carried out to assess the effect of noisy measurements on the estimated parameters.

• Journal of Mechanical Science and Technology
• /
• v.17 no.1
• /
• pp.85-96
• /
• 2003

In the present investigation, the nonlinear dynamic buckling of a curved plate subjected to sinusoidal loading is examined. By the theoretical analyses, a highly nonlinear snap-through motion of a clamped-free-clamped-free plate and its effect on the overall vibration response are investigated. The problem is reduced to that of a single degree of freedom system with the Rayleigh-Ritz procedure. The resulting nonlinear governing equation is solved using Runge-Kutta (RK-4) numerical integration method. The snap-through boundaries, which vary with different damping coefficient and linear circular frequency of the flat plate are studied and given in terms of force and displacement. The relationships between static and dynamic responses at the start of a snap-through motion are also predicted. The analysis brings out various characteristic features of the phenomenon, i.e. 1) small oscillation about the buckled position-softening spring type motion, 2) chaotic motion of intermittent snap-through, and 3) large oscillation of continuous snap-through motion crossing the two buckled positions-hardening spring type. The responses of buckled plate were found to be greatly affected by the snap-through motion. Therefore, better understanding of the snap-through motion is needed to predict the full dynamic response of a curved plate.