• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.1
• /
• pp.19-26
• /
• 2023

Because of the development of computational resources, an entire structure in which many components are combined can be analyzed. To do so, the calculation time and number of data points are increased. In many practical industry structures, there are many parts with repeated patterns. To analyze the repetitive structures effectively, a homogenization method is usually employed. In a homogenization module, including commercial programs, it is generally assumed that a unit cell is repeated in all directions. However, the practical industry structures usually have complicated, repeated patterns or structures. Complicated patterns are difficult to address using the conventional homogenization method. Therefore, in this study, a multilevel homogenization method was devised to consider complex regularities. The proposed homogenization method divides the structure into several areas and performs multiple homogenizations, resulting in a more accurate analysis than that provided by the previous method.

• Proceedings of the Safety Management and Science Conference
• /
• 2011.11a
• /
• pp.67-74
• /
• 2011

In this study numerical analysis was performed to evaluate the safety of Built-in Timbering Roof Type Tunnelling Method(BTR) which is one of non-opening tunnel construction methods. For the upgrading of analytical precision was applied the discretion method which can separately model reinforcing elements of BTR and the homogeneity analysis considering the area ratio of elements was performed to compare both results. Comparing the displacement in this study with that of the homogeneity method, the efficiency of the discretion method was verified.

• Composites Research
• /
• v.36 no.3
• /
• pp.205-210
• /
• 2023

In this study, we developed a transfer learning framework based on homogenization data for efficient prediction of the effective mechanical properties and thermal conductivity of cellular foam structures. Mean-field homogenization (MFH) based on the Eshelby's tensor allows for efficient prediction of properties in porous structures including ellipsoidal inclusions, but accurately predicting the properties of cellular foam structures is challenging. On the other hand, finite element homogenization (FEH) is more accurate but comes with relatively high computational cost. In this paper, we propose a data-driven transfer learning framework that combines the advantages of mean-field homogenization and finite element homogenization. Specifically, we generate a large amount of mean-field homogenization data to build a pre-trained model, and then fine-tune it using a relatively small amount of finite element homogenization data. Numerical examples were conducted to validate the proposed framework and verify the accuracy of the analysis. The results of this study are expected to be applicable to the analysis of materials with various foam structures.

• Journal of Soil and Groundwater Environment
• /
• v.9 no.1
• /
• pp.79-86
• /
• 2004

Hydraulic conductivity along rock fracture is mainly dependent on fracture geometries such as orientation, aperture, roughness and connectivity. Therefore, it needs to consider fracture geometries sufficiently on a fracture model for a numerical analysis to calculate permeability coefficient in a fracture. This study performed new type of numerical analysis using a homogenization analysis method to calculate permeability coefficient accurately along single fractures with several fracture models that were considered fracture geometries as much as possible. First of all, fracture roughness and aperture variation due to normal stress applied on a fracture were directly measured under a confocal laser scaning microscope (CLSM). The acquired geometric data were used as input data to construct fracture models for the homogenization analysis (HA). Using the constructed fracture models, the homogenization analysis method can compute permeability coefficient with consideration of material properties both in microscale and in macroscale. The HA is a new type of perturbation theory developed to characterize the behavior of a micro inhomogeneous material with a periodic microstructure. It calculates micro scale permeability coefficient at homogeneous microscale, and then, computes a homogenized permeability coefficient (C-permeability coefficient) at macro scale. Therefore, it is possible to analyze accurate characteristics of permeability reflected with local effect of facture geometry. Several computations of the HA were conducted to prove validity of the HA results compared with the empirical equations of permeability in the previous studies using the constructed 2-D fracture models. The model can be classified into a parallel plate model that has fracture roughness and identical aperture along a fracture. According to the computation results, the conventional C-permeability coefficients have values in the range of the same order or difference of one order from the permeability coefficients calculated by an empirical equation. It means that the HA result is valid to calculate permeability coefficient along a fracture. However, it should be noted that C-permeability coefficient is more accurate result than the preexisting equations of permeability calculation, because the HA considers permeability characteristics of locally inhomogeneous fracture geometries and material properties both in microscale and macroscale.

• Composites Research
• /
• v.36 no.5
• /
• pp.329-334
• /
• 2023

The classical multiscale finite element (FE² ) method involves iterative calculations of micro-boundary value problems for representative volume elements at every integration point in macro scale, making it a computationally time and data storage space. To overcome this, we developed the data-driven multiscale analysis method based on the mean-field homogenization (MFH). Data-driven computational mechanics (DDCM) analysis is a model-free approach that directly utilizes strain-stress datasets. For performing multiscale analysis, we efficiently construct a strain-stress database for the microstructure of composite materials using mean-field homogenization and conduct data-driven computational mechanics simulations based on this database. In this paper, we apply the developed multiscale analysis framework to an example, confirming the results of data-driven computational mechanics simulations considering the microstructure of a hyperelastic composite material. Therefore, the application of data-driven computational mechanics approach in multiscale analysis can be applied to various materials and structures, opening up new possibilities for multiscale analysis research and applications.

• Tunnel and Underground Space
• /
• v.19 no.2
• /
• pp.158-166
• /
• 2009

As a basic study for investigating the development of the stress-induced crack in Hoek-Brown rock, a homogenization technique of elastic cracks is proposed. The onset of crack is monitored by Hoek-Brown empirical criterion, while the orientation of the crack is determined by the critical plane approach. The concept of volume averaging in stress and strain component was invoked to homogenize the representative rock volume which consists of intact rock and cracks. The formulation results in the constitutive relations for the homogenized equivalent anisotropic material. The homogenization model was implemented in the standard FEM code COSMOSM. The numerical uniaxial tests were performed under plane strain condition to check the validity of the propose numerical model. The effect of friction between the loading plate and the rock sample on the mode of deformation and fracturing was examined by assuming two different contact conditions. The numerical simulation revealed that the homogenized model is able to capture the salient features of deformation and fracturing which are observed commonly in the uniaxial compression test.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.21 no.1
• /
• pp.13-20
• /
• 2008

Graded layers in which two different constituent particles are mixed are inserted into functionally graded material such that the volume fractions of constituent particles vary continuously and functionally over the entire material domain. The material properties of this dual-phase graded region, which is essential for the numerical analysis of the thermo-mechanical behavior of FGM, have been predicted by traditional homogenization methods. But, these methods are limited to predict the global equivalent material properties of FGMs because the detailed geometry information such as the particel shape and the dispersion structure is not considered. In this context, this study intends to investigate the characteristics of these homogenization methods through the finite element analysis utilizing the discrete micromechanics models of the graded layer, for various volume fractions and external loading conditions.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.25 no.4
• /
• pp.315-321
• /
• 2012

In this study, a sequential multiscale homogenization method to characterize the effective thermal conductivity of nano particulate polymer nanocomposites is proposed through a molecular dynamics(MD) simulations and a finite element-based homogenization method. The thermal conductivity of the nanocomposites embedding different-sized nanoparticles at a fixed volume fraction of 5.8% are obtained from MD simulations. Due to the Kapitza thermal resistance, the thermal conductivity of the nanocomposites decreases as the size of the embedded nanoparticle decreases. In order to describe the nanoparticle size effect using the homogenization method with accuracy, the Kapitza interface in which the temperature discontinuity condition appears and the effective interphase zone formed by highly densified matrix polymer are modeled as independent phases that constitutes the nanocomposites microstructure, thus, the overall nanocomposites domain is modeled as a four-phase structure consists of the nanoparticle, Kapitza interface, effective interphase, and polymer matrix. The thermal conductivity of the effective interphase is inversely predicted from the thermal conductivity of the nanocomposites through the multiscale homogenization method, then, exponentially fitted to a function of the particle radius. Using the multiscale homogenization method, the thermal conductivities of the nanocomposites at various particle radii and volume fractions are obtained, and parametric studies are conducted to examine the effect of the effective interphase on the overall thermal conductivity of the nanocomposites.

• The Journal of Engineering Geology
• /
• v.20 no.3
• /
• pp.271-279
• /
• 2010

Time-dependent behavior similar to secondary deformation related to mineral dissolution is easily observed when performing a laboratory pressure experiment. In this research, to observe the dissolution of quartz found in bentonite used as buffer material for the geological disposal of high-level waste (HLW) under conditions of high pH, we calculated the diffusion of $OH^-$ ions and the behavior of quartz dissolution using the homogenization analysis method. The results reveal that the rate of quartz dissolution is proportional to the temperature and interlayer water thickness. In particular, in a high-pH environment, the reacted area (and therefore the dissolution rate) increases with decreasing interlayer water thickness.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2011.04a
• /
• pp.261-264
• /
• 2011

유한 요소 기법을 이용한 허니콤 구조물의 해석은 모델링 작업 및 격자 생성의 어려움뿐만 아니라 과도한 해석 시간이 요구되기 때문에 균질화 기법은 계산상의 효율성을 증대시킬 수 있는 매우 유용한 방법이라 할 수 있다. 그러나 나노 크기의 구조물에서는 표면 효과로 인하여 거시적인 구조물에서와는 매우 상이한 기계적 거동 양상을 띠게 되며 균질화 기법을 나노 크기의 허니콤 구조물에 적용하기 위해서는 이러한 표면 효과를 반영해야만 한다. 본 논문에서는 표면 효과를 고려한 유한 요소를 제안하고 이를 이용하여 나노 크기의 3차원 허니콤 구조물을 균질화 기법을 이용하여 등가의 2차원 판으로 대체하였다.