• Geomechanics and Engineering
• /
• 제34권6호
• /
• pp.683-696
• /
• 2023

The destruction and fracture of rock masses are crucial components in engineering and there is an increasing demand for the study of the influence of rock mass heterogeneity on the safety of engineering projects. The numerical manifold method (NMM) has a unified solution format for continuous and discontinuous problems. In most NMM studies, material homogeneity has been assumed and despite this simplification, fracture mechanics remain complex and simulations are inefficient because of the complicated topology updating operations that are needed after crack propagation. These operations become computationally expensive especially in the cases of heterogeneous materials. In this study, a heterogeneous model algorithm based on stochastic theory was developed and introduced into the NMM. A new fracture algorithm was developed to simulate the rupture zone. The algorithm was validated for the examples of the four-point shear beam and semi-circular bend. Results show that the algorithm can efficiently simulate the rupture zone of heterogeneous rock masses. Heterogeneity has a powerful effect on the macroscopic failure characteristics and uniaxial compressive strength of rock masses. The peak strength of homogeneous material (with heterogeneity or standard deviation of 0) is 2.4 times that of heterogeneous material (with heterogeneity of 11.0). Moreover, the local distribution of parameter values can affect the configuration of rupture zones in rock masses. The local distribution also influences the peak value on the stress-strain curve and the residual strength. The post-peak stress-strain curve envelope from 60 random calculations can be used as an estimate of the strength of engineering rock masses.

• Nuclear Engineering and Technology
• /
• 제53권9호
• /
• pp.2968-2976
• /
• 2021

This paper introduces a method to reconstruct the three-dimensional (3D) microstructure of two-phase materials, e.g., porous materials such as highly irradiated nuclear fuel, from two-dimensional (2D) sections via a multi-objective optimization genetic algorithm. The optimization is based on the comparison between the reference and reconstructed 2D sections on specific target properties, i.e., 2D pore number, and mean value and standard deviation of the pore-size distribution. This represents a multi-objective fitness function subject to weaker hypotheses compared to state-of-the-art methods based on n-points correlations, allowing for a broader range of application. The effectiveness of the proposed method is demonstrated on synthetic data and compared with state-of-the-art methods adopting a fitness based on 2D correlations. The method here developed can be used as a cost-effective tool to reconstruct the pore structure in highly irradiated materials using 2D experimental data.

• Coupled systems mechanics
• /
• 제8권4호
• /
• pp.289-298
• /
• 2019

Reinforced concrete can be considered as a heterogeneous material consisting of coarse aggregate, mortar mix and reinforcing bars. This paper presents a two-dimensional mesoscopic analysis of reinforced concrete beams using a simple two-phase mesoscopic model for concrete. The two phases of concrete, coarse aggregate and mortar mix are bonded together with reinforcement bars so that inter force transfer will occur through the material surfaces. Monte Carlo's method is used to generate the random aggregate structure using the constitutive model at mesoscale. The generated models have meshed such that there is no material discontinuity within the elements. The proposed model simulates the load-deflection behavior, crack pattern and ultimate load of reinforced concrete beams reasonably well.

• Computers and Concrete
• /
• 제26권1호
• /
• pp.11-20
• /
• 2020

Concrete are the most widely used manmade materials for infrastructure construction across the world. These constructions gradually aged and damaged due to long-term use. However, there does not exist an efficient concrete recycling method with low energy consumption. In this study, concrete was regarded as a heterogeneous material composed of coarse aggregate and cement mortar. And the failure mode of concrete under ultrasonic dynamic loading was investigated by finite element (FE) analysis. Simultaneously, a 3D random aggregate concrete model was programmed by APDL and imported into ABAQUS software, and the damage plastic constitutive model was applied to each phase to study the damage law of concrete under dynamic loading. Meanwhile, the dynamic damage process of concrete was numerically simulated, which observed ultrasonic propagating and the concrete crushing behavior. Finally, the FE simulation considering the influence of different aggregate volume and aggregate size was carried out to illustrate the damage level of concrete.

• International Journal of Concrete Structures and Materials
• /
• 제18권1E호
• /
• pp.23-28
• /
• 2006

In this study, the general theory of fractality is discussed to provide a fundamental understanding of fractal geometry applied to heterogeneous material surfaces like pavement surface and rock surface. It is well known that many physical phenomena and systems are chaotic, random and that the features of roughness are found at a wide spectrum of length scales from the length of the sample to the atomic scales. Studying the mechanics of these physical phenomena, it is absolutely necessary to characterize such multi scaled rough surfaces and to know the structural property of such surfaces at all length scales relevant to the phenomenon. This study emphasizes the role of fractal geometry to characterize the roughness of various surfaces. Pavement roughness and rock surface roughness were examined to correlate their roughness property to fractality.

• Computers and Concrete
• /
• 제5권3호
• /
• pp.175-194
• /
• 2008

Increasingly numerical (finite element) modeling of concrete hinges on our ability to develop a representative volume element with all its heterogeneity properly discretized. Yet, despite all the sophistication of the ensuing numerical models, the initial discretization has been for the most part simplistic. Whenever the heterogeneity of the concrete is to be accounted for, a mesh is often manually crafted through the arbitrary inclusion of the particles (aggregates and/or voids) in an ad-hoc manner. This paper develops a mathematical strategy to precisely address this limitation. Algorithms for the random generation and placement of elliptical (2D) or ellipsoid (3D) inclusions, with possibly radiating cracks, in a virtual concrete model are presented. Collision detection algorithms are extensively used.

• Investigative Magnetic Resonance Imaging
• /
• 제2권1호
• /
• pp.58-66
• /
• 1998

목적: 본 논문은 마코브 랜덤필드(Markov Random Field)와 깁스 랜덤필드(Gibbs Random Field) 및 라인모델(LIne Model)에 기반한 3차원 자기공명영상의 분류 방법을 소개하고자 하였다. 대상 및 방법 : 통계적으로 이질적 성분들로 구성된 영상을 대상으로한 깁스분류 결과는 영상내의 국소적으로 정적인 영역들을 이웃화소 시스템 내에서 정의되는 상호작용 인자(inetraction parameter)의 메커니즘에 의해 분리하\ulcorner로서 개선시킬 수 있다. 이를 위하여 영상에서 라인모델의 생성을 고려할 수 있으며, 본 논문에서는 영상의 미분방법에 근거한 다중신호영상을 위한 라인모델을 구축하였다. 라인모델은 서로 상이한 통게적 특성을 갖는 영역사이에 존재하는 관측할 수 없는 라인필드의 존재 유무를 확률 값으로 제공한다. 영상으로부터 획득한 라이모델은 Gibbs 분류기의 에너지함수 값을 결정하는 상호작용 인자 값을 결정하는데 사용된다. 결과 : 3차원 자기공명영상의 분류를 위한 MRS-Gibbs 분류기는 영상분류의 도메인이 일반적인 이차원 영상의 $E^{2}$ 공간에서 $E^{3}$ 공간으로 확장되었다. 개발된 깁스분류기를 이용한 자기공명여상의 분류결과 기존의 context free 분류방법에 의한 결과에 비하여 특히 동일성질을 갖고 있는 영역 및 경계부분 등의 분류결과가 우수함을 알 수 있었다. 결론 : 본 논문에서는 다중 신호, 3차원 자기공명영상을 위한 라인모델을 구축하고 그로부터 MRF-Gibbs분류기의 에너지함수를 결정하기 위한 상호작용 인자를 유도하였다.

• Interaction and multiscale mechanics
• /
• 제3권1호
• /
• pp.23-37
• /
• 2010

Concrete is a heterogeneous material consisting of coarse aggregate, mortar matrix and interfacial zones at the meso level. Though studies have been done to interpret the fracture process in concrete using meso level models, not much work has been done for simulating the macroscopic behaviour of reinforced concrete structures using the meso level models. This paper presents a procedure for the mesoscopic analysis of reinforced concrete beams using a modified micro truss model. The micro truss model is derived based on the framework method and uses the lattice meshes for representing the coarse aggregate (CA), mortar matrix, interfacial zones and reinforcement bars. A simple procedure for generating a random aggregate structure is developed using the constitutive model at meso level. The study reveals the potential of the mesoscopic numerical simulation using a modified micro truss model to predict the nonlinear response of reinforced concrete structures. The modified micro truss model correctly predicts the load-deflection behaviour, crack pattern and ultimate load of reinforced concrete beams failing under different failure modes.

• 마이크로전자및패키징학회지
• /
• 제30권4호
• /
• pp.32-43
• /
• 2023

저항 변화 메모리 소자(RRAM)는 저항 변화 특성을 기반으로 빠른 동작 속도, 간단한 소자 구조 및 고집적 구조의 구현을 통해 많은 양의 데이터를 효율적으로 처리할 수 있는 차세대 메모리 소자로 주목받고 있다. RRAM의 작동원리 중 하나로 알려진 interface type의 저항 변화 특성은 forming process를 수반하지 않고 소자 크기를 조절하여 낮은 전류에서 구동이 가능하다는 장점을 갖는다. 그 중에서도 전이 금속 산화물 기반 RRAM 소자의 경우, 정확한 물질의 조성 조절 방법과 소자의 신뢰성 및 안정성과 같은 메모리 특성을 향상시키기 위해 다양한 연구가 진행 중에 있다. 본 논문에서는 이종 원소의 도핑, 다층 박막의 형성, 화학적 조성 조절 및 표면 처리 등의 방법을 이용하여 interface type 저항 변화 특성의 저하를 방지하고 소자 특성을 향상시키기 위한 다양한 방법을 소개하고자 한다. 이를 통해 향상된 저항 변화 특성을 기반으로 한 고효율 차세대 비휘발성 메모리 소자의 구현 가능성을 제시한다.

• Asian-Australasian Journal of Animal Sciences
• /
• 제33권11호
• /
• pp.1741-1754
• /
• 2020

Objective: This study was conducted to estimate genetic parameters for milk yield traits using daily milk yield records from parlour data generated in an intensively managed commercial dairy farm with Jersey and Jersey-Friesian cows in Sri Lanka. Methods: Genetic parameters were estimated for first and second lactation predicted and realized 305-day milk yield using univariate animal models. Genetic parameters were also estimated for total milk yield for each 30-day intervals of the first lactation using univariate animal models and for daily milk yield using random regression models fitting second-order Legendre polynomials and assuming heterogeneous residual variances. Breeding values for predicted 305-day milk yield were estimated using an animal model. Results: For the first lactation, the heritability of predicted 305-day milk yield in Jersey cows (0.08±0.03) was higher than that of Jersey-Friesian cows (0.02±0.01). The second lactation heritability estimates were similar to that of first lactation. The repeatability of the daily milk records was 0.28±0.01 and the heritability ranged from 0.002±0.05 to 0.19±0.02 depending on day of milk. Pearson product-moment correlations between the bull estimated breeding values (EBVs) in Australia and bull EBVs in Sri Lanka for 305-day milk yield were 0.39 in Jersey cows and -0.35 in Jersey-Friesian cows. Conclusion: The heritabilities estimated for milk yield in Jersey and Jersey-Friesian cows in Sri Lanka were low, and were associated with low additive genetic variances for the traits. Sire differences in Australia were not expressed in the tropical low-country of Sri Lanka. Therefore, genetic progress achieved by importing genetic material from Australia can be expected to be slow. This emphasizes the need for a within-country evaluation of bulls to produce locally adapted dairy cows.