Alain Gasser

Analyses of fabric tensile behaviour: determination of the biaxial tension–strain surfaces and their use in forming simulations

The forming of fibre fabric reinforcements without a matrix is possible because of their very specific mechanical behaviour. The lack of some rigidities is due to possible motions between the fibres. For the fabrics used as reinforcement in the R.T.M. process and composed of warp and weft yarns made with untwisted fibres, the tension stiffness is very preponderant compared to the others. The tensile behaviour of such a fabric is biaxial, i.e. the tension-deformation states in warp or weft directions depend on the other direction because of the interweaving. It is given by the knowledge of two surfaces relating the warp and weft tensions to the two strains in these directions (or that of a single surface if the fabric is balanced). In the present paper, three complementary methods are investigated in order to determine these surfaces. A biaxial tensile device on a cross-shaped specimen is first used. 3D finite element simulations of the unit woven cell are then presented. This mesoscopic study permits to understand some phenomena at the elementary woven cell level. Finally a simplified model, which is consistent with the geometry of the plain weave woven mesh is presented. The agreement of the two last methods with experimental results is shown. From these tensile behaviour surfaces, a dynamic explicit approach for the simulations of a fabric sheet forming process is presented. The interests of the method are both its good numerical efficiency, particularly due to the direct use of the biaxial tension surfaces, and its proximity with fabric physics.

Damage evolution in [±45]s laminates with fiber rotation

Abstract The competing effects of increasing stiffness due to fiber rotation and decreasing stiffness due to damage evolution are incorporated into the Ladeveze mesoscale damage model to study the large strain response of [±45]s laminates of IM7-K3B, graphite-fiber/polyimide-matrix composite. It is shown that both of these competing effects are significant and the lack of their consideration can result in significant error when predicting large strain, non-linear response. Predictions of the model with and without fiber rotation are compared with experimental results for the tensile response of [±45]s laminates. It is also shown that variations in cross-sectional area of test specimens plays an important role in the large strain response of this high Poisson-ratio laminate. Improved correlation between theory and experiment is demonstrated when all competing effects are included in the analysis. The study includes predictions of stresses and damage variables in the individual layers of the laminate as a function of the modeling approach.

Mechanical Behavior of Woven Composite Reinforcements While Forming

Simulation of the forming processes of thin composite structures is necessary at the design level in order to check the feasibility of the shape and to know the position of the reinforcements. A finite element analysis of fabric shaping process needs the knowledge of the mechanical behavior of the woven reinforcement. This behavior is non linear because of the shape variation of the weaving pattern when it is loaded. Experimental results are obtained from biaxial tests in the case of balanced and unbalanced fabric. A constitutive model consistent with the geometry of the woven pattern is proposed. It is based on experimental results achieved by biaxial tensile tests. 3D simulations of the unit woven cell submitted to biaxial tensions are also performed and compared to experiments.

Computations of refractory lining structures under thermal loadings

Refractory linings are used to protect the exterior metallic part of some vessels containing very hot fluids. They are submitted to high thermomechanical loading that can lead to cracking. A local approach is first presented in order to analyse the refractory lining as a 3D domain. A smeared crack model is used to compute the damage in the refractory. Comparison with experiments on a refractory wall containing metal parts is performed in order to validate the 3D numerical computations. Some type of refractorised vessels (e.g. some steel ladles) can directly be analysed from this 3D modelling. Since some other refractorised vessel contains a very large number of metallic parts (such as tubes or anchors), it cannot be possible to compute such a global structure with this 3D analysis. Consequently, an approach has been developed based on a two-layer shell equivalent to the lining including the metallic casing with tubes and the refractory. The thermal and mechanical parameters of the model are identified with an inverse method, using results of 3D calculations performed with the local model defined previously. An experimental validation is made by a bending test, performed on a large refractory lining specimen. In the case of a cyclone of coal-fired power plant, the equivalent shell permits to compute the damage of the refractory in the global structure.

Thermomechanical behaviour analysis and simulation of steel/refractory composite linings

In steel ladles (steel industry) or in coal-fired power plants (energy production), the refractory linings anchored to the steel structure (casing) are subjected to significant thermal loadings which lead to cracking (owing to the difference between the thermal expansion coefficients). A first approach allows us to analyse the problem at a local scale: a smeared crack model permits the computation of the damage which is observed in an experimental test. But since a structure contains thousand of anchors, it is not possible to compute it with this model. An approach was therefore developed with a two-layered shell equivalent to the anchored lining and the casing. This model is identified with an inverse method and uses results from the first approach, validated by a bending test performed on a refractory lining specimen. The damage computing of a structure shows that it is very important to take the expansion joints into account.

Simulations of thermomechanical behavior of composite refractory linings

The wall of refractorized vessels are composites, made of metallic casing often containing tubes and a refractory material for the protection against the high temperature environment. The objective of the present paper is to model a two layer shell with a thermomechanical behavior equivalent to those of the 3D lining and that render possible the finite element analysis of the complete vessel. A smeared crack model is used for the damage analysis of the refractory material. The equivalent shell is made up of an exterior orthotropic layer and an interior isotropic damageable layer. The set of thermal and mechanical parameters of the equivalent shell is obtained by an inverse method in conjunction with finite element analyses of the 3D panel subjected to an appropriate set of loadings. Some validation analyses show that the identified parameters lead to shell behavior, which is in good agreement with those of the 3D wall. In the case of a simplified cyclone, it is shown that the equivalent shell permits to compute the thermomechanical behavior of a complete refractorized vessel and especially to follow the damaged zones of the refractory.

Modelling of joint effects on refractory lining behaviour

Abstract Expansion joints play an important role in refractory linings as they reduce stresses during heating. It is therefore necessary to take them into account in a mechanical analysis. In the case where the lining is a masonry construction (made up of bricks), it would require an excessive number of elements to model each brick and joint. The proposed solution is to replace the masonry with a material that has the same behaviour (or very near) as masonry. Since it is difficult to perform experimental tests on a set of bricks (to identify the parameters of the equivalent material), these loads were simulated on an elementary cell using a model developed at a local scale (scale of the components, bricks and joints). At this scale, the joints are represented as contacts (with normal and tangential behaviour). The parameters of a simplified equivalent material were obtained by an inverse identification. This model was validated by a thermomechanical test on a real structure.

Modeling of Coal Drying before Pyrolysis

The coking process is composed of two main stages: drying process and pyrolysis of coal. A heat and mass transfer model was developed to simulate the drying process of coal. The mechanisms of heat and mass transfer described in the model are: conduction through the coal cake; conduction and convection through the gas in pores; generation, flux and condensation of water vapor. The model has been implemented in finite element software. It requires basic data on the coke oven charge properties and oven dimensions as input. These input data were obtained by experiments or from the literature. The proposed model includes condensation and evaporation allowing us to reproduce the temperature plateau observed experimentally.

HOMOGENIZATION TECHNIQUES FOR ACCURATE AND APPROXIMATE ESTIMATES FOR OVERALL PROPERTIES OF MICROCRACKED VISCOELASTIC MASONRIES

Based on the coupling between brittle fracture mechanics and homogenization techniques, either analytical or finite elements homogenizatio n, this work provides accurate (up to numerical errors) and approximate overall estimates for ma sonries accounting for their creep behaviour and a certain level of damage as it is the case for in sta ce for refractory linings serving at high temperatures or middle-ages masonry buildi ng.

Multiphysics Modelling Applied to Refractory Behaviour in Severe Environments

It is a common practice to design refractory linings with the help of thermal computations, thermochemistry analyses and strong workman know-how. Their mechanical design is often limited to simple thermo-elastic computations. Sometimes computations are refined considering non-linear mechanical behaviour, even if corrosion often induces additional chemical strain and strong change in service of the mechanical behaviour of the refractory. The aim of this presentation is to briefly recast the irreversible thermodynamic framework in order to underline the implications of some basic thermodynamic concepts in term of refractory behaviour modelling. Then, the use of these concepts to develop fully 3D finite element simulations accounting simultaneously for thermal, mechanical and chemistry phenomena will be illustrated on the particular case of SiC-based refractory. Comparison between long duration oxidation test at high temperature and model prediction allows the validation of the proposed approach. Then, an extension to the industrial case of refractory lining in Waste to Energy plant will be illustrated. The interest of taking into account the thermo-chemo-mechanical coupling effects is shown.