Filipe Teixeira-Dias
University of Aveiro
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Filipe Teixeira-Dias.
Computational Materials Science | 2001
Filipe Teixeira-Dias; L.F. Menezes
Abstract It is natural to suppose that some of the technological factors associated to the processes used in the fabrication of metal matrix composite (MMC) materials can and will influence in some extent the performance of these materials when in service. This is often true due to the levels of residual stresses that may be induced in the MMC after the cooling down phase during the fabrication process. In the present work, the authors propose a complete three-dimensional constitutive model and numerical implementation procedure that allows the determination of residual stress fields in metal matrix composites. The model is based in a thermoelastic reinforcement behaviour and a thermoelastic–viscoplastic matrix behaviour. The role of the reinforcement volume fraction and cooling rate on the levels of residual stresses at room temperature is investigated with the proposed model. For this purpose, a large set of simulations is performed with Al–SiC metal matrix composites. Two different unit cells are used, representative of continuous and short fiber reinforcement MMCs. The tested reinforcement volume fractions range from 5% to 35% and cooling rates from 0.1 to 500 K s−1. The influence of these parameters is evaluated in terms of the resulting stress fields at room temperature. It is shown that the levels of equivalent stress can reach values above the yield limit of the aluminium matrix, leading to plastic straining near the matrix/reinforcement interface.
Holzforschung | 2007
Celina Pires Gameiro; José Cirne; Victor Miranda; J. Pinho-da-Cruz; Filipe Teixeira-Dias
Abstract Cork is a unique and complex natural cellular material with many industrial applications. The purpose of this paper is to explore a new application field for the use of micro-agglomerate cork as an energy-absorbing medium. A numerical study on the energy absorption capabilities of square and circular cork-filled aluminium tubes with a width or diameter of 80 mm, length of 300 mm and variable thickness was performed with the finite element method code LS-DYNA™. The tubes were impacted uniaxially at 10 and 15 m s-1. The same analysis was carried out on aluminium foam-filled tubes. The results demonstrate that cork filling leads to a considerable increase in the energy absorbed for both section geometries, and that tube thickness plays an important role in the deformation modes and energy absorption. The investigation revealed better results for aluminium foam-filled structures, but demonstrated that micro-agglomerate cork has high potential as an energy-absorbing medium in crash protection applications.
Computational Materials Science | 2002
Filipe Teixeira-Dias; L.F. Menezes
Abstract A kinematic model suitable for the numerical analysis of dual-phase materials and, in particular, for composite materials is presented. Due to the non-linear character of the behavior of most materials, the model must be presented in an incremental form. The kinematic model is based on the decomposition of the transformation gradient in thermoelastic and plastic parts. Thus, the velocity gradient tensor is also decomposed in an elastic and a plastic part. Two different constitutive models are implemented in order to test the model. The dual-phase material modeled is an Al–SiC metal matrix composite (MMC). The reinforcement constitutive model is chosen to be thermoelastic and the matrix model is thermoelastic–viscoplastic. A forward gradient time integration procedure is described and implemented in order to calculate the increments of the state variables, namely the plastic strain increment and the stress increments in both materials. The proposed model is tested with one three-dimensional example: a cylindrical fiber MMC with 11% reinforcement volume fraction. The numerical results are compared with results obtained by other authors. Very good accordance can be observed between the results obtained with the incremental model proposed in this paper and those presented by other authors.
Key Engineering Materials | 2013
R. M. F. Paulo; Pierpaolo Carlone; R. A. F. Valente; Filipe Teixeira-Dias; Gaetano Salvatore Palazzo
Stiffened panels are usually the basic structural building blocks of airplanes, vessels and other structures with high requirements of strength-to-weight ratio. They typically consist of a plate with equally spaced longitudinal stiffeners on one side, often with intermediate transverse stiffeners. Large aeronautical and naval parts are primarily designed based on their longitudinal compressive strength. The structural stability of such thin-walled structures, when subjected to compressive loads, is highly dependent on the buckling strength of the structure as a whole and of each structural member. In the present work, a number of modelling and numerical calculations, based on the Finite Element Method (FEM), is carried out in order to predict the ultimate load level when stiffened panels are subjected to compressive solicitations. The simulation models account not only for the elasto-plastic nonlinear behaviour, but also for the residual stresses, material properties modifications and geometrical distortions that arise from Friction Stir Welding (FSW) operations. To construct the model considering residual stresses, their distribution in FSW butt joints are obtained by means of a numerical-experimental procedure, namely the contour method, which allows for the evaluation of the normal residual stress distribution on a specimen section. FSW samples have been sectioned orthogonally to the welding line by wire electrical discharge machining (WEDM). Displacements of the relaxed surfaces are then recorded using a Coordinate Measuring Machine and processed in a MATLAB environment. Finally, the residual stress distribution is evaluated by means of an elastic FE model of the cut sample, using the measured and digitalized out-of-plane displacements as input nodal boundary conditions. With these considerations, the main goal of the present work will then be related to the evaluation of the effect of FSW operations, in the ultimate load of stiffened panels with complex cross-section shapes, by means of realist numerical simulation models.
Simulation | 2012
Ricardo J. Alves de Sousa; Daniel Gonçalves; Rodrigo Coelho; Filipe Teixeira-Dias
The efficiency of cork as a material dedicated to energy absorption under impact loading is studied in the present work. The viability of the application of micro-agglomerate cork (MAC) padding on a motorcycle helmet, is studied using finite element simulations of impact tests, considering the specifications of the European Standard ECE-R.22/05. Expanded polystyrene (EPS) is a widely used material, with excellent results in energy-absorption applications. However, after a first impact, the capability of EPS for energy absorption is significantly decreased, due to the almost total absence of elastic springback. However, cork is a material characterized by having both a good energy-absorption capability and high elastic return, due to its viscoelastic behavior, meaning that its capacity to keep absorbing energy is almost unchanged after the first impact. In this work, a three-dimensional numerical model of the helmet–head system is developed, including the outer shell, safety padding and the head, together with its interactions and constitutive models suitable for the analyzed materials. Results show that the developed models can adequately reproduce the behavior of EPS and MAC, in the context of a preliminary analysis. The referred helmet–headform is then submitted to impacts at different points, as specified by the European Standard. The results from helmeted impacts with EPS padding are compared against experimental values. The application of MAC in the protective padding of the helmet is studied and the results, concerning the acceleration of the gravity center of the head, Head Injury Criterion (HIC) values and kinetic energy are presented. Results obtained with EPS and MAC are compared and discussed. Concerning cork, although the maximum acceleration values of the headform and the HIC values were not verified to be within the established limits of the regulatory standard, the results are promising, launching a sound basis for a more thorough work on the application of cork as a new material for advanced applications as an energy-absorption system.
Archive | 2010
Filipe Teixeira-Dias; L.F. Menezes
It is well known that residual stresses strongly influence the behaviour of most materials and, in particular, of composite materials. This chapter presents one approach to the numerical determination of thermal residual stresses in metal matrix composites (MMC). The subject of residual stresses is introduced and the corresponding mathematical and constitutive models are described in detail. It is considered that the reinforcement material is elastic and that the metallic matrix may exhibit thermoelastic-viscoplastic behaviour. A progressive gradient based time-integration algorithm is described that leads to the implementation of the proposed constitutive models in a finite element analysis code. The corresponding variational formulation and discretisation into finite elements is also described. In order to guarantee stabilised convergence and to increase the precision of results, the authors also propose a time-step optimisation algorithm. All the formalisms are tested measuring the influence of the reinforcement volume fraction and cooling rate on the resulting residual stresses.
Archive | 2008
J. A. Oliveira; J. Pinho-da-Cruz; Filipe Teixeira-Dias
Finite element (FE) simulation plays a crucial role in the analysis of the mechanical behaviour of structural elements built with complex microstructure composite materials. In order to define microstructural details, finite element analysis (FEA) often leads to the need for unstructured meshes and large numbers of finite elements. This fact frequently makes it impossible to perform numerical analyses on the mechanical behaviour of such structural components, due to the large amounts of required memory and CPU time. In this particular context, homogenisation methodologies lead to significant computational benefits.
Materials Science Forum | 2007
A. Andrade-Campos; Filipe Teixeira-Dias
Residual stress fields can cause creep damage in thermally aged components, even in the absence of working loads. In order to study this issue, the authors present a numerical study on the development of triaxial residual stresses in stainless steel specimens. A mechanical model dedicated to the analysis of heat treatment problems is described. The presented formulations are implemented incrementally with a non-linear constitutive model, adequate to the simulation of a wide range of thermal processes. The flow rule is a function of the equivalent stress and the deviatoric stress tensor, of the temperature field and of a set of internal state variables. The thermomechanical coupled problem is solved with a staggered approach. Spray water quenching was used to generate residual stress fields in solid cylinders and spheres made from 316H stainless steel. Finite element simulations were performed to find out how process conditions and specimen geometry influence the resulting residual stress distributions. The results show that compressive residual stresses are developed near the surfaces of the cylinders and spheres while tensile residual stresses occur near the centre. The level of residual stresses was found to be dependent on the heat transfer coefficient.
Key Engineering Materials | 2012
R. M. F. Paulo; Filipe Teixeira-Dias; R. A. F. Valente
Stiffened panels are composed of a base plate with stiffeners in one or more directions, leading to lightweight structures with high resistance. The structural design, in most cases, focuses mainly on the longitudinal compressive loads that the panels are subjected and can safely withstand. In the present work, a set of Finite Element Method Analyses (FEA) were carried out, using ABAQUS commercial simulation software, and compared with experimental data in order to infer about the sensitivity of the results to the initial geometrical imperfections (either in magnitude and shape). The developments in the present work aim to provide a range of models able to properly reproduce the experimental behaviour of aluminium stiffened panels subjected to compressive loads. It was shown that FEA using shell finite elements were able to obtain accurate predictions of the ultimate load, considering large deformation and elasto-plastic behaviours. The effect of using different shapes and magnitudes of the initial geometrical imperfections on the numerical simulation of the panels was also inferred and tested using previously obtained eigenvalue (EV) buckling modes.
Archive | 2011
João A. Oliveira; J. Pinho-da-Cruz; Filipe Teixeira-Dias
Composite materials are among the most prominent materials today, both in terms of applications and development. Nevertheless, their complex structure and heterogeneous nature lead to difficulties, both in the prediction of its properties and on the achievement of the ideal constituent distributions. Homogenisation procedures may provide answers in both cases. With this in mind, the main focus of this chapter is to show the importance of computational procedures for this task, mainly in terms of the different applications of Asymptotic Expansion Homogenisation (AEH) to heterogeneous periodic media and, above all, composite materials. First of all, it is noteworthy that the detailed numerical modelling of the mechanical behaviour of composite material structures tends to involve high computational costs. In this scope, the use of homogenisation methodologies can lead to significant benefits. These techniques allow the simplification of a heterogeneous medium using an equivalent homogenousmedium andmacrostructural behaviour laws obtained from microstructural information. Furthermore, composite materials typically have heterogeneities with characteristic dimensions significantly smaller than the dimensions of the structural component itself. If the distribution of the heterogeneities is roughly periodic, it can usually be approximated by a detailed periodic representative unit-cell. Thus, the Asymptotic Expansion Homogenisation (AEH) method is an excellent methodology to model physical phenomena on media with periodic microstructure, as well as a useful technique to study the mechanical behaviour of structural components built with compositematerials. In terms of computational implementation, the main advantages of this method are (i) the fact that it allows a significant reduction of the number of degrees of freedom and (ii) the capability to find the stress and strain microstructural fields associated with a given macrostructural equilibrium state. In fact, unlike other common homogenisation methods, the AEH leads to explicit mathematical equations to characterise those fields, that is, to perform a localisation. On the other hand, topology optimisation typically deals with material distributions to achieve the best behaviour for a given objective. The common approach to structural topology optimisation uses a variety of compliance minimisation (stiffness maximisation) procedures and functions. When analysing composite materials, these strategies often lead to multiscale procedures, either as a way to relax the initial discrete problem or in an effort to attain both optimal global structure and optimal microstructure. In this sense, the integration of AEH 23