• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2001년도 가을 학술발표회 논문집
• /
• pp.19-26
• /
• 2001

Previously, an improved crack analysis technique based on Element-Free Galerkin Method (EFGM) which includes a discontinuity function and a singular basis function was presented. The technique needs neither addition of nodes nor modification of the model, but it shows some dependency on the formulation and modeling parameters such as the class of weight function, the size of compact support, dilation parameter and the range controlled by the singular basis function. For those parameters, a parametric study was performed on the calculation of a discrete error and then, a guideline for the choice of adequate parameters in the technique was proposed.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1992년도 가을 학술발표회 논문집
• /
• pp.7-19
• /
• 1992

An efficient approach to limit analysis is presented whereby a continuous perfectly plastic structure is replaced by a discrete mathematical model. It is formulated as a mathematical programming problem using the static theorem of plasticity. The discretization is accomplished by writing the governing equilibrium equations in finite difference form, and is combined with piecewise linearization of the nonlinear yield curve, thus converting the formulation into a linear programming exercise. Examples of reported cases involving plates and shells are solved to illustrate the ease of application of the present method, its flexibility and accuracy - features which it make attractive to practising engineers.

• 대한기계학회논문집
• /
• 제17권10호
• /
• pp.2457-2466
• /
• 1993

A simple method is presented for determining transient responses of locally nonlinear structures using substructure eigenproperties and Lagrange multiplier technique. Although the method is based upon the mode synthesis formulation procedure, the equations of the combined whole structure are not constructed compared with the conventional methods. Lagrange multi-pliers are used to enforce the conditions of geometric compatibility between the substructure interfaces and they are treated as external forces on each substructure itself. Substructure eigenvalue problem is defined with the substructure interface free of fixed. The transient analysis is based upon the recurrence discrete-time state equations and offers the simplicity of the Euler integration method without requiring small time increment and iterative solution procedure. Numerical examples reveal that the method is very accurated and efficient in calculating transient responses compared with the direct numerical integration method.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1995년도 하계학술대회 논문집 B
• /
• pp.804-806
• /
• 1995

This paper discusses Self-organized Distributed Networks(SODN) as identifier of nonlinear dynamical systems. The structure of system identification employs series-parallel model. The identification procedure is based on a discrete-time formulation. The learning with the proposed SODN is fast and precise. Such properties arc caused from the local learning mechanism. Each local networks learns only data in a subregion. Large number of memory requirements and low generalization capability for the untrained region, which are drawbacks of conventional local network learning, are overcomed in the SODN. Through extensive simulation, SODN is shown to be effective for identification of nonlinear dynamical systems.

• Structural Engineering and Mechanics
• /
• 제54권4호
• /
• pp.665-691
• /
• 2015

This study aims at comparing the optimum design of two common types open web expanded beams: with hexagonal openings, also called castellated beams and beams with circular openings referred to as cellular beams. The minimum weights of both beams are taken as the objective functions while the design constraints are respectively implemented from The Steel Construction Institute Publication Numbers 5 and 100. The design methods adopted in these publications are consistent with BS5950 parts. The formulation of the design problem considering the limitations of the above mentioned turns out to be a discrete programming problem. Improved harmony search algorithm is suggested to compare the optimum design of mentioned web-expanded beams to analysis the performance of both beams. The design algorithms based on the technique select the optimum Universal Beam sections, dimensional properties of hexagonal and circular holes and total number of openings along the beam as design variables.

• 대한전기학회논문지
• /
• 제45권4호
• /
• pp.574-581
• /
• 1996

The neural network approach has been shown to be a general scheme for nonlinear dynamical system identification. Unfortunately the error surface of a Multilayer Neural Networks(MNN) that widely used is often highly complex. This is a disadvantage and potential traps may exist in the identification procedure. The objective of this paper is to identify a nonlinear dynamical systems based on Self-Organized Distributed Networks (SODN). The learning with the SODN is fast and precise. Such properties are caused from the local learning mechanism. Each local network learns only data in a subregion. This paper also discusses neural network as identifier of nonlinear dynamical systems. The structure of nonlinear system identification employs series-parallel model. The identification procedure is based on a discrete-time formulation. Through extensive simulation, SODN is shown to be effective for identification of nonlinear dynamical systems. (author). 13 refs., 7 figs., 2 tabs.

• 전기학회논문지
• /
• 제61권7호
• /
• pp.1019-1023
• /
• 2012

This paper considers a model predictive control problem of discrete-time constrained systems with time-varying delay. For this problem, a delay dependent state feedback control approach is used to achieve asymptotic stabilization of systems with input constraints. Based on Lyapunov stability theory, a new stability condition is obtained via linear matrix inequality formulation to find cost monotonicity condition of the model predictive control algorithm which guarantee the closed loop stability. Finally, the proposed method is applied to a numerical example in order to show the effectiveness of our results.

• 한국도로학회논문집
• /
• 제15권5호
• /
• pp.31-45
• /
• 2013

PURPOSES : Determination of particle packing model variables that can be used for formulation of new DEM based particle packing model by examining existing particle packing models METHODS : Existing particle packing models are thoroughly examined by analytical reformulation and sensitivity analysis in order to set up DEM based new particle packing model and to determine its variables. All model equations considered in this examination are represented with consistent expressions and are compared to each others to find mathematical and conceptual similarity in expressions. RESULTS : From the examination of existing models, it is observed that the models are very similar in their shapes although the derivation of the models may be different. As well, it is observed that variables used in some existing models are comprehensive enough to estimate particle packing but not applicable to DEM simulation. CONCLUSIONS : A set of variables that can be used in DEM based particle packing model is determined.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2004년도 추계학술대회논문집
• /
• pp.469-475
• /
• 2004

The transient analysis of delaminated smart composite laminates is studied using an improved layerwise laminate theory. The theory is capable of capturing interlaminar shear stresses that are critical to delamination. The presence of discrete delamination is modeled through the use of Heaviside unit step functions. Stress free boundary conditions are enforced at all fee surfaces. Continuity in displacement field and transverse shear stresses are enforced at each ply level. In modeling piezoelectric composite plates, a coupled piezoelectric-mechanical formulation is used in the development of the constitutive equations. Numerical analysis is conducted to investigate the effect of nonlinearity in the transient vibration of bimodular behavior caused by the contact impact of delaminated interfaces. Composite plates with delamination, subject to external loads and voltage history from surface bonded sensors, are investigated and the results are compared with corresponding experimental results and plates without delamination.

• Structural Engineering and Mechanics
• /
• 제55권6호
• /
• pp.1157-1176
• /
• 2015

This study presents optimizing structural topology patterns using regularization of Heaviside function. The present method needs not filtering process to typical SIMP method. Using the penalty formulation of the SIMP approach, a topology optimization problem is formulated in co-operation, i.e., couple-signals, with design variable values of discrete elements and a regularized Heaviside step function. The regularization of discontinuous material distributions is a key scheme in order to improve the numerical problems of material topology optimization with 0 (void)-1 (solid) solutions. The weak forms of an equilibrium equation are expressed using a coupled regularized Heaviside function to evaluate sensitivity analysis. Numerical results show that the incorporation of the regularized Heaviside function and the SIMP leads to convergent solutions. This method is tested using several examples of a linear elastostatic structure. It demonstrates that improved optimal solutions can be obtained without the additional use of sensitivity filtering to improve the discontinuous 0-1 solutions, which have generally been used in material topology optimization problems.