Begell House Inc.
International Journal for Multiscale Computational Engineering
JMC
1543-1649
12
1
2014
INVESTIGATION OF MICRO-MACROSCALE INTERACTION OF HETEROGENEOUS MATERIALS BY A PARALLEL-BONDED PARTICLE MODEL AND INTRODUCTION OF NEW MICROPARAMETER DETERMINATION FORMULATIONS
1-21
Serkan
Nohut
Zirve University, Faculty of Engineering, Kizilhisar Kampusu, 27260, Gaziantep, Turkey
Abdulkadir
Cevik
University of Gaziantep, Department of Civil Engineering, 27310, Gaziantep, Turkey
The distinct element method (DEM) is becoming an effective method of investigating engineering problems in granular and heterogeneous materials, especially in granular flows, powder mechanics, advanced ceramics, and rock mechanics. Creation of a DEM model requires some microscale material parameters, unable to be physically measured in laboratories; a calibration process is typically used in order to select the proper microparameters using DEM simulations. The calibration process is basically trial and error, which depends on the experience of the modeler. Therefore such a process may be complicated and time consuming for the user. In this study, a parametric study is performed in order to determine the relations of microparameters used in three dimensional DEM model and the macroscale material parameters (i.e., Young's modulus, Poisson's ratio, compressive strength). According to dependencies and independencies between microparameters and macroparameters, empirical fitting functions are obtained by using a stepwise regression method. The macroparameters calculated by empirical fitting functions reveal a good agreement with DEM results. The predictive ability of fitting functions is confirmed with the creation of further data sets in DEM simulations. Comparison of the fitting values with the literature shows that the fitting functions may also be used two dimensional DEM simulations. The micro-macro scale interactions and empirical fitting functions provided in this study would be very helpful for the user in order to observe the relationship between micro- and macroparameters and determine the approximate proper microparameters.
MODELING OF TWO-PHASE GRAIN STRUCTURE IN THE TITANIUM ALLOY TI-6AL-4V USING CELLULAR AUTOMATA
23-31
Alfred
Krumphals
Institute for Materials Science and Welding, Christian Doppler Laboratory for Materials Modelling and Simulation, Graz University of Technology, Graz, Austria
Cecilia
Poletti
Institute for Materials Science and Welding, Christian Doppler Laboratory for Materials Modelling and Simulation, Graz University of Technology, Graz, Austria
Fernando
Warchomicka
Institute of Materials Science and Technology, Vienna University of Technology, Vienna, Austria
Martin
Stockinger
Bohler Schmiedetechnik GmbH & Co KG, Kapfenberg, Austria
Christof
Sommitsch
Institute for Materials Science and Welding, Christian Doppler Laboratory for Materials Modelling and Simulation, Graz University of Technology, Graz, Austria
The static coarsening behavior of the alpha-beta titanium alloy Ti-6Al-4V during heat treatments is modeled using a probabilistic cellular automata model (CA). For this purpose the kinetics of grain growth is described via transformation probabilities which are determined by diffusion mechanisms at grain and phase boundaries. For temperature changes an algorithm is implemented which adjusts the fraction of alpha and beta phase to reach equilibrium phase values. Hence, the CA is capable of calculating grain coarsening as well as grain dissolution in the two-phase area during heating and isothermal treatments at forging temperature. For these calculations, an initial microstructure is used as input and it can be imported from either virtual created microstructures, real micrographs, or electron backscatter diffraction (EBSD) maps. The model output includes mean diameter, grain size distribution, and virtually simulated microstructures which can be easily compared with experimental micrographs. Examples showing a good correlation between the predicted microstructures and experimental results, as well as data from literature, are presented in this work. The successful implementation of this model will lead to predictions of behavior in other dual-phase alloys.
BOUNDARY ELEMENT METHOD MODELLING OF NANOCOMPOSITES
33-43
Jacek
Ptaszny
Institute of Computational Mechanics and Engineering, Silesian University of Technology, Konarskiego 18A, 44-100 Gliwice, Poland
Grzegorz
Dziatkiewicz
Institute of Computational Mechanics and Engineering, Silesian University of Technology, Konarskiego 18A, 44-100 Gliwice, Poland
Piotr
Fedelinski
Institute of Computational Mechanics and Engineering, Silesian University of Technology, Konarskiego 18A, 44-100 Gliwice, Poland
The paper deals with the numerical homogenization of polymer/clay nanocomposites reinforced by stacks of parallel clay sheets. The stacks can be modelled as effective particles, as it was shown in the literature. For a relatively small volume fraction of the reinforcement, the effective particles can be isotropic, while for greater values, the particles should be anisotropic. Other authors most commonly use analytical methods or the finite element method (FEM). In this work, the boundary element method (BEM) is applied. Two-dimensional plain strain models are analyzed. Two cases are considered, namely, isotropic and anisotropic (orthotropic) particles. The matrix of the composite is modelled as isotropic. The problem is solved by using a BEM formulation for plates containing many identical inclusions. The kernels of boundary integrals for the isotropic subdomains are the Kelvin solutions for plane elasticity. In the case of the orthotropic particles, fundamental solutions obtained by the Stroh formalism are applied. The results are compared to the Mori-Tanaka model. Acceptable agreement between the models is observed.
REDUCED-ORDER MULTISCALE-MULTIPHYSICS MODEL FOR HETEROGENEOUS MATERIALS
45-64
Zheng
Yuan
Multiscale Design Systems LLC, 280 Park Ave, Apt 22M, New York, NY 10010, U.S.A.
Tao
Jiang
Multiscale Design Systems LLC, 280 Park Ave, Apt 22M, New York, NY 10010, U.S.A.
Jacob
Fish
Civil Engineering and Engineering Mechanics, Columbia University, New York, New York
10027, USA
Greg
Morscher
Department of Mechanical Engineering, University of Akron, Akron, OH 44325, U.S.A.
A unified coupled multiscale mechano-diffusion-reaction model of environmental degradation of polymer matrix composite (PMC) and ceramic matrix composite (CMC) is developed. The unified multiscalemultiphysics model couples multiple physical processes at multiple scales, including oxygen diffusion, oxidation, and deformation. The salient feature of the unified multiscalemultiphysics model is its computational efficiency accomplished through a systematic model reduced carried out prior to nonlinear analysis. The model has been validated for PMR-15 reinforced carbon fiber composite and melt infiltrated CMC-NASA N24A material system.
SENSITIVITY ANALYSIS OF TRANSIENT TEMPERATURE FIELD IN MICRODOMAINS WITH RESPECT TO THE DUAL-PHASE-LAG MODEL PARAMETERS
65-77
Ewa
Majchrzak
Institute of Computational Mechanics and Engineering, Silesian University of Technology, ul. Konarskiego 18a, 44-100 Gliwice, Poland
Bohdan
Mochnacki
Czestochowa University of Technology, Dabrowskiego 69, 42-201 Czestochowa, Higher School of Labour Safety Management, Bankowa 8, 40-007 Katowice, Poland
In the paper selected problems concerning microscale heat-transfer modeling are presented. In particular, the dual-phase-lag model (DPLM) containing two time lags, the relaxation and thermalization times, is considered. The aim of this research is to estimate the changes in the transient temperature field due to the perturbations of the DPLM thermophysical parameters (volumetric specific heat, thermal conductivity, and relaxation and thermalization times). To solve the problem methods of sensitivity analysis (direct approach) are applied. At the stage of numerical modeling, the axially symmetrical object subjected to an external heat flux is considered. Numerical computations are realized using the explicit scheme of finite difference method. In the final part of the paper the examples of computations are shown and the conclusions are formulated.
BIOINSPIRED IDENTIFICATION OF PARAMETERS IN MICROSCALE HEAT TRANSFER
79-89
Jolanta
Dziatkiewicz
Faculty of Mechanical Engineering, Institute of Computational Mechanics and Engineering,
Silesian University of Technology, Poland
Waclaw
Kus
Institute of Computational Mechanics and Engineering, Silesian University of Technology, ul. Konarskiego 18a, 44-100 Gliwice, Poland
Ewa
Majchrzak
Institute of Computational Mechanics and Engineering, Silesian University of Technology, ul. Konarskiego 18a, 44-100 Gliwice, Poland
Tadeusz
Burczynski
Institute of Fundamental Technological Research, Polish Academy of Sciences
Lukasz
Turchan
Faculty of Mechanical Engineering, Institute of Computational Mechanics and Engineering,
Silesian University of Technology, Poland
The paper is devoted to the identification of microscale heat-transfer parameters. The numerical modeling of short-pulse laser interaction with thin metal films is considered. The hyperbolic two-temperature model describing the temporal and spatial evolution of the lattice and electrons temperatures in the irradiated metal is applied. This model consists of four equations: two equations concern the electron and lattice temperatures; the later ones determine the dependencies between heat fluxes and temperatures. The short-pulse laser interaction with the film is taken into account by introducing an internal volumetric heat source to the equation describing the electron temperature. The equations concerning the electrons and lattice temperatures are joined by coupling factor G, which characterizes the energy exchange between phonons and electrons. The relations between electron heat flux and electron temperature and between the lattice heat flux and lattice temperature contain the parameters ?e and ?l, respectively. The parameter ?e is the relaxation time of free electrons in metals; the parameter ?l is the relaxation time in phonon collisions. The one-dimensional problem is analyzed. (Heat transfer in the direction perpendicular to the thin film is taken into account.) The nonflux conditions can be accepted at the front surface irradiated by a laser pulse and the back surface. The initial conditions are also assumed. The direct problem is solved by the explicit scheme of the finite difference method. The results of the computations are partially compared with the experimental data available in literature. The inverse problem discussed here consists in the simultaneous identification of three parameters, namely, the coupling factor G and relaxation times ?e and ?l. To solve such a problem, the electron temperature history at the irradiated surface of the thin film is taken into account. The inverse problems can be formulated as optimization problems and solved by means of bioinspired algorithms. The objective function is formulated on the basis of the known measured and numerical simulated values of temperature. The minimization of the objective function allows one to find the design variables vector, which may contain the parameters of the coupling factor and time coefficients in the presented case. The inverse problems are ill-defined problems, and the identification may lead to different results with the same objective function value. The objective function can have many local minima, and therefore the bioinspired algorithm is used in the paper.