• Transactions of the Korean Society of Mechanical Engineers
• /
• v.17 no.10
• /
• pp.2543-2554
• /
• 1993

In this paper, the kinematic and dynamic analyses of spatial complex kinematic chain are studied. Through the new method both using the set of identification numbers and applying the DenavitHartenberg link representation method to the spatial complex kinematic chain, the kinematic configuration of the chain is represented. Some link in the part of closed chain being fictitiously cutted, the complex kinematic chain is transformed to the branched chain. The kinematic constraint equations are derived from the constraint conditions which the cutted sections of the link have to satisfy. And the joint variables being partitioned in the independent joint variables and the dependent joint variables, the dependent variables are calculated from the independent variables by using the Newton-Raphson iterative method and the pseudoinverse matrix. The equations of motion are derived under the independent joint variables by using the principle of virtual work. Algorithms for dynamic analysis are presented and simulations are done to verify accuracy and efficiency of the algorithms.

• Structural Engineering and Mechanics
• /
• v.55 no.6
• /
• pp.1241-1260
• /
• 2015

Any rational approach to define the configuration and size of viscous fluid dampers in a structure should be based on the dynamic properties of the system with the dampers. In this paper we propose an alternative representation of the complex eigenvalues of multi degree of freedom systems with dampers to calculate new equivalent natural frequencies. Analytical expressions for the dynamic properties of a two-story building model with a linear viscous damper in the first floor (i.e. with a non-proportional damping matrix) are derived. The formulas permit to obtain the equivalent damping ratios and equivalent natural frequencies for all the modes as a function of the mass, stiffness and damping coefficient for underdamped and overdamped systems. It is shown that the commonly used formula to define the equivalent natural frequency is not applicable for this type of system and for others where the damping matrix is not proportional to the mass matrix, stiffness matrix or both. Moreover, the new expressions for the equivalent natural frequencies expose a novel phenomenon; the use of viscous fluid dampers can modify the vibration frequencies of the structure. The significance of the new equivalent natural frequencies is expounded by means of a simulated free vibration test. The proposed approach may offer a new perspective to study the effect of viscous dampers on the dynamic properties of a structure.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.8
• /
• pp.2061-2074
• /
• 1998

In pattern recognition and image analysis upplications, a graph is a useful tool for complex obect representation and recognition. However it takes much time to pair proper nodes between the prototype graph and an input data graph. Futhermore it is difficult to decide whether the two graphs in a class are the same hecause real images are degradd in general by noise and other distortions. In this paper we propose a matching algorithm using a matrix. The matrix is suiable for simple and easily understood representation and enables the ordering and matching process to be convenient due to its predefined matrix manipulation. The nodes which constitute a gaph are ordered in the matrix by their geometrical positions and this makes it possible to save much comparison time for finding proper node pairs. for the classification, we defined a distance measure thatreflects the symbo's structural aspect that is the sum of the mode distance and the relation distance; the fornet is from the parameters describing the node shapes, the latter from the relations with othes node in the matrix. We also introduced a subdivision operation to compensate node merging which is mainly due t the prepreocessing error. The proposed method is applied to the recognition of musteal symbols and the result is given. The result shows that almost all, except heavily degraded symbols are recognized, and the recognition rate is approximately 95 percent.

• Journal of the military operations research society of Korea
• /
• v.14 no.1
• /
• pp.42-52
• /
• 1988

Many programmers start debugging by reading the faulty program from start to bottom without investigating carefully the erroneous program. Expert programmers, however, trace backward from a particular variable in a specific statement to identify all possible sources of influence on the value of variable (program slice). Weiser proposed a slicing algorithm (method) that is complex, iterative and still in modification [3,4]. This paper presents a method to generate a program slice by use of matrix computation which represents all possible slices of the program. The matrix representation of a program is soundly based on the graph theory of data dependency.

• Journal of the military operations research society of Korea
• /
• v.13 no.2
• /
• pp.42-52
• /
• 1987

Many programmers start debugging by reading the faulty program from start to bottom without investigating carefully the erroneous program. Expert programmers, however, trace backward from a particular variable in a specific statement to identify all possible sources of influence on the value of variable (program slice). Weiser proposed a slicing algorithm (method) that is complex, iterative and still in modification [3,4]. This paper presents a method to generate a program slice by use of matrix computation which represents all possible slices of the program. The matrix representation of a program is soundly based on the graph theory of data dependency.

• Korean Journal of Mathematics
• /
• v.27 no.2
• /
• pp.375-415
• /
• 2019

We study spaces of continuous-function-valued functions that have the property that composition with evaluation functionals induce $weak^*$ to norm continuous maps to ${\ell}^p$ space ($p{\in}(1,\;{\infty})$). Versions of $H{\ddot{o}}lder^{\prime}s$ inequality and Riesz representation theorem are proved to hold on these spaces. We prove a version of Dixmier's theorem for spaces of function-valued matrix operators on these spaces, and an analogue of the trace formula for operators on Hilbert spaces. When the function space is taken to be the complex field, the spaces are just the ${\ell}^p$ spaces and the well-known classical theorems follow from our results.

• Journal of KIISE:Software and Applications
• /
• v.32 no.8
• /
• pp.789-794
• /
• 2005

LDA (Linear Discriminant Analysis) provides the projection that discriminates the data well, and shows a very good performance for face recognition. However, since LDA provides only one transformation matrix over whole data, it is not sufficient to discriminate the complex data consisting of many classes like honan faces. To overcome this weakness, we propose a new face recognition method, called LDA mixture model, that the set of alf classes are partitioned into several clusters and we get a transformation matrix for each cluster. This detailed representation will improve the classification performance greatly. In the simulation of face recognition, LDA mixture model outperforms PCA, LDA, and PCA mixture model in terms of classification performance.

• Geomechanics and Engineering
• /
• v.2 no.4
• /
• pp.229-252
• /
• 2010

In this paper, Soil-Structure Interaction (SSI) effect is investigated using a new and integrated approach. Faster solution of time dependant differential equation of motion is achieved using numerical representation of wavelet theory while dynamic Infinite Elements (IFE) concept is utilized to effectively model the unbounded soil domain. Combination of the wavelet theory with IFE concept lead to a robust, efficient and integrated technique for the solution of complex problems. A direct method for soil-structure interaction analysis in a two dimensional medium is also presented in time domain using the frequency dependent transformation matrix. This matrix which represents the far field region is constructed by assembling stiffness matrices of the frequency dependant infinite elements. It maps the problem into the time domain where the equations of motion are to be solved. Accuracy of results obtained in this study is compared to those obtained by other SSI analysis techniques. It is shown that the solution procedure discussed in this paper is reliable, efficient and less time consuming as compared to other existing concepts and procedures.

• Structural Engineering and Mechanics
• /
• v.43 no.2
• /
• pp.147-161
• /
• 2012

This paper presents a curvature method for analysis of beam-columns with different materials and arbitrary cross-section shapes and subjected to combined biaxial moments and axial load. Both material and geometric nonlinearities (the p-delta effect in this case) were incorporated. The proposed method considers biaxial curvatures and uniform normal strains of discrete cross-sections of beam-columns as basic unknowns, and seeks for a solution of the column deflection curve that satisfies force equilibrium conditions. A piecewise representation of the beam-column deflection curve is constructed based on the curvatures and angles of rotation of the segmented cross-sections. The resulting bending moments were evaluated based on the deformed column shape and the axial load. The moment curvature relationship and the beam-column deflection calculation are presented in matrix form and the Newton-Raphson method is employed to ensure fast and stable convergence. Comparison with results of analytic solutions and eccentric compression tests of wood beam-columns implies that this method is reliable and effective for beam-columns subjected to eccentric compression load, lateral bracings and complex boundary conditions.

• Korean Journal of Optics and Photonics
• /
• v.18 no.6
• /
• pp.410-420
• /
• 2007

We theoretically derive the set of utilizable paraxial zoom locus equations for all complicated zoom lens systems with infinite object distance, such as a camera zoom lens, by using the Gaussian bracket method and the matrix representation of paraxial ray tracing. And we make the zoom locus program according to these equations in Visual Basic. Since we have applied the paraxial ray tracing equations into Gaussian bracket representation, the resultant program systematically simplifies various constraints of the zoom loci of various N group types. Consequently, the solutions of this method can be consistently used in all types of zoom lens in the step of initial design about zoom loci. Finally, in order to verify the usefulness of this method, we show that one example among 4 groups and that among 5 groups, which are very complex zoom lens systems, can be rapidly and with versatility traced through various interpolations by using this program.