• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.29 no.5
• /
• pp.463-469
• /
• 2016

In this study, a virtual specimen with a linear continuous gradient of void ratio (FGM: Functional Graded Material) is constructed using low-order probability functions of two real cement paste specimens. Two real specimens with difference void ratios are taken from X-ray CT to construct the virtual specimen. A virtual specimen with a gradient void distribution, whose average void ratio is between void ratios of two homogeneous real specimens, is constructed using a stochastic optimization approach. The void ratio distribution is assumed to be linear, and continuously varies in the vertical direction. In this study, a gradient term of void ratio is incorporated into the objective function as well as low-order probability functions from the previous research. To confirm the effect of gradient void distribution on the material response, air permeability is evaluated using finite element analysis. The analysis results are compared with experimental results, and confirm the effect of gradient void distribution on permeability.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.5A
• /
• pp.425-431
• /
• 2010

The phase distribution in a multi-phase material strongly affects its material properties. Therefore, a proper method to describe the phase distribution of a material is needed. In this research, probability distribution functions, two-point correlation and lineal-path functions, are used to represent the probabilistic phase distributions of a material. The probability distribution function is calculated using a numerical method and is described as an analytical form via exponential curve fitting with three parameters. Application of analytical form of probability distribution function is investigated using two-phase polycrystalline solids and soil samples. It is confirmed that the probability distribution functions can be represented as an exponential form using curve fitting which helps identifying the applicability of a representative volume element(RVE).

• Composites Research
• /
• v.21 no.4
• /
• pp.7-14
• /
• 2008

Prediction of the load-bearing capacity of an adhesive-bonded Joint is of practical importance for engineers. This paper introduces interface fracture mechanics approach to predict the load-bearing capacity of composite metal bonded joints. The adhesion strength of composite/steel bonding is evaluated in terms of the energy release rate of an interfacial crack and the fracture toughness of the interface. Virtual track closure technique (VCCT) is used to calculate energy release rates, and hi-material end-notched flexure (ENF) specimens are devised to measure the interfacial fracture toughness. Bi-material ENF specimens gave consistent mode II fracture toughness $(G_{IIc})$ values of the composite/steel interface regardless of the thickness of specimens. The critical energy release rates of double-lap joints showed a good agreement with the measured fracture toughness. Therefore. the energy-based interfacial fracture characterization can be a practical engineering tool for predicting the load-bearing capacity of bonded joints.

• Composites Research
• /
• v.18 no.4
• /
• pp.44-51
• /
• 2005

In this paper, we developed the direct numerical simulation(DNS) model considering the geometry of yams which consist of 3D orthogonal woven composite materials, and using this model, the dynamic behavior of under transverse low-velocity impact has been studied. To build up the micromechanical model considering tow spacing and waviness, an accurate unit structure is presented and used in building structural plate model based on DNS. For comparison, DNS results are compared with those of the micromechanical approach which is based on the global equivalent material properties obtained by DNS static numerical tests. The effects with yarn geometrical irregularities which are difficult to consider in a macroscopic approach are also investigated by the DNS model. Finally, the multiscale model based on the DNS concepts is developed to enhance efficiency of analysis with real sized numerical specimen and macro/micro characteristics are presented.

• Magazine of the Korea Concrete Institute
• /
• v.11 no.1
• /
• pp.119-127
• /
• 1999

The objective of this study is to simulate the fracture behavior of composite structure bonded with more than 2 different cementitious materials. For this, concrete and cement were stacked and bonded in a direction perpendicular to loading and specimens were tested. Each constituent material of concrete and cement was fabricated independently also, and three point bending and indirect tensile tests were carried out for the acquisition of measured values applicable to the proposed model. As a result of comparing theoretical results and experimental ones, it was found that the proposed model derived from fictitious crack theory can be used to predict the fracture behavior of composite structures on the vases of well agreement with experimental results. It was also noted that the degree of improvement of fracture energies and strengths is greatly dependent on the stacking sequence of layers composing of a composite structure. Thus, it can be concluded that brittleness or ductility of a composite structure can be accomplished by a proper arrangement of layers on one's purpose throughout the proposed analysis.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.34 no.5
• /
• pp.293-300
• /
• 2021

In this study, the microstructure and thermal conductivity correlation was investigated for concrete materials used in concrete thermal energy storage (CTES) among real-time energy storage devices. Graphite was used as admixture to increase the thermal conductivity performance of the CTES. Concrete specimens of 10% and 15% substitution of cement by mass with graphite, as well as ordinary portland cement (OPC) specimens were prepared, and the microstructural changes and effects on thermal conductivity were analyzed. Porosities of OPC and concrete with graphite were compared using micro-CT, and the microstructural characteristics were quantified using probability functions. Three-dimensional virtual specimens were constructed for thermal analysis, to confirm the effect of microstructural characteristics on thermal conductivity, and the results were compared with the measured conductivity obtained using the hot-disc method. To identify thermal conductivity of graphite for thermal analysis, solid phase conductivity was inversely determined based on simulation and experimental results, and the effect of graphite on thermal conductivity was analyzed.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.31 no.7
• /
• pp.41-52
• /
• 2003

In this paper, three-dimensional thermo-mechanical properties of carbon-phenolic 8-hamess satin weave composites were predicted considering geometric parameters of microstructures. The effective properties were calculated by a series of numerical experiments based on unit cell analysis. The microstructural details were modeled through macro-elements, and the periodic boundary conditions were derived for corresponding un it cell types. The Monte Carlo method was employed to consider the random phase shift between the layers, and the results were investigated on the effect of the geometric parameters of shift, number of layers and waviness ratios. Experimental tests were also performed and the results were compared.