Search

This study aims at better understanding the damage and fracture kinetics in flax fibre elements at both the unitary and bundle scales, using an experimental setup allowing optical observation at high recording rate in the course of tensile loading. Defects and issues from flax unitary fibre extraction are quantitated using polarized light microscopy. Tensile loading is conducted according to a particular setup, adapted to fibres of 10 to 20 µm in diameter and 10 mm in length. Optical recording using a high-speed camera is performed during loading up to the failure at acquisition, with speed ranging from 108,000 to 270,000 frames per second. Crack initiation in polymer layers of fibre elements, propagation as well as damage mechanisms are captured. The results show different failure scenarios depending on the fibre element’s nature. In particular, fractured fibres underline either a fully transverse failure propagation or a combination of transverse and longitudinal cracking with different balances. Image recordings with high time resolution of down to 3.7 μs suggest an unstable system and transverse crack speed higher than 4 m/s and a slower propagation for longitudinal crack deviation. Failure propagation monitoring and fracture mechanism studies in individual natural fibre or bundles, using tensile load with optical observation, showed contrasted behaviour and the importance of the structural scale exanimated. This study can help in tailoring the eco-design of flax-based composites, in terms of toughness and mechanical performances, for both replacement of synthetic fibre materials and innovative composites with advanced properties.

This paper addresses the complexity reduction of stochastic homogenization of a class of random materials for a stationary diffusion equation. A cost-efficient approximation of the correctors is obtained using a method designed to exploit quasi-periodicity. Accuracy and cost reduction are investigated for local perturbations or small transformations of periodic materials as well as for materials with no periodicity but a mesoscopic structure, for which the limitations of the method are shown. Finally, for materials outside the scope of this method, we propose to use the approximation of homogenized quantities as control variates for variance reduction of a more accurate and costly Monte Carlo estimator (using a multifidelity Monte Carlo method). The resulting cost reduction is illustrated in a numerical experiment and compared with a control variate method from weakly stochastic homogenization. The limits of this variance reduction technique are tested on materials without periodicity or mesoscopic structure.

Fibrous textiles are subjected to dry through-thickness compaction during most of the composites manufacturing processes. Usual tow models implemented in textile numerical simulations do not reproduce the tow widening occurring during the compaction. Nonetheless, this widening (that can reach 10%) would influence the internal microstructure of the considered fabric and its mechanical behavior. This paper proposes a simple mechanical approach to reproduce the width and thickness evolutions experimentally measured during through-thickness compaction of E-glass and carbon tows. Once the material parameters identified on laterally free tows, the tow cross-section model is employed to study the influence of the tow lateral confinement on the transverse mechanical behavior of its corresponding quasi-unidirectional fabric. It is observed that the lateral confinement depends on the quasi-unidirectional stitch tension. This confinement induces a densification of the tows leading to a stiffening of the quasi-unidirectional transverse behavior. This stiffening is well predicted by the proposed modeling approach.

© 2022 Elsevier. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/ This paper presents a new methodology to extend the Data-Driven Identification (DDI) to heterogeneous samples made of multiple elastic materials. By using the Correspondence Analysis (CA) technique to post-process DDI, we are able to identify multiple material databases representative of the material behavior of each phase. Simultaneously, we localize the different phases (matrix and inclusions) in the sample. For different contrasts between phases, the method is tested on synthetically generated data and a parametric study is performed. Furthermore, we show that it is possible to iterate between DDI and CA in order to improve the method’s predictions when it is limited by scarce input data. In all the cases studied, the methodology proves to be effective for estimating stresses, as well as for identifying the different phases in the sample. This project is part of the Marie Sklodowska-Curie ITN-EJD ProTechTion funded by the European Union Horizon 2020 research and innovation program with Grant Number 764636.

For fast and complete impregnation in Liquid Composite Molding, knowledge about the permeability of the fibrous reinforcement is required. While development of experimental methods continues, a parallel benchmark effort to numerically characterize permeability is being pursued. The approach was to send out the images of a real fibrous microstructure to a number of participants, in order for them to apply their methods for virtual permeability prediction. Via resin transfer molding a plate was manufactured, using the glass woven fabric Hexcel 01102 (295 g/m²) at a fiber volume content of 54% and a thermoset resin. From this plate, a specimen was scanned using a 3D x-ray microscope at a scan size of 1000 x 1000 x 1000 μm³ and a resolution of 0.521 μm³ per voxel. The sample extracted for the simulations with a size of 523 x 65 x 507 μm³ contains about 400 fibers of a single tow. It revealed a variation of filament diameters between 7.5-9.3 μm and a fiber volume content in average of 56.46% with a variation of 54 - 59% in the individual 2D-slices transverse to the fiber direction. The image segmentation was performed by 2D-slices, to which a Hough transform was applied to detect fiber centers and cross-sections. Then fiber paths were tracked through-out the slices by the closest neighbor algorithm. Finally, fiber paths were smoothened by means of the local regression using weighted linear least squares and a 1st degree polynomial model. The participants received a stack of 973 segmented (binary) 2D-images and a corresponding segmented 3D volume raw-file. They were asked to calculate the full permeability tensor components and fill out a detailed questionnaire including questions e.g. on applied flow models and conditions, numerical discretization and approximation methods, fluid properties etc. The received results scatter considerably over two orders of magnitude, although the participants were provided an already segmented image structure, thus eliminating from the beginning a significant source of variation that could have come from image processing. Model size, meshing and many other sources of variation were identified, allowing further specification of the guidelines for the next step.

“Quilted Stratum Process” (QSP®) is a new process in the category of thermoplastic composite forming with the objective to locally strengthen the composite parts by strategically stacking discontinuous UD/woven prepregs while maintaining the short cycle time of about 1 min. Interply adhesion arising due to polymer tack plays an important role in QSP® due to the presence of resin-rich layer at the ply-ply interface, inability to use blank holders for prepreg patches and high temperature of forming process where the resin is in melt state. Without modeling interply adhesion in the numerical simulation of their forming, plies delaminate unrealistically without any resistance which in turn results in incorrect final positions of the prepreg patches. Thus, a penalty based, semi-empirical contact model for interply adhesion has been developed and implemented in the industrial finite element code of Altair RADIOSS™. This model allows sliding of plies over long distance while providing a finite adhesive strength before delamination. It requires minimal characterization for which a measurement method is proposed. The usage of this model in a full scale numerical simulation showed that the final ply positions were predicted with a much better accuracy. Also, the predictions of fibre orientations within individual plies are in good agreement with the experimental observations.

The mechanical efficiency of fiber-reinforced composite materials mainly depends on the alignment of fibers along the load paths arising from the boundary conditions applied to the part. On this basis, the Tailored Fiber Placement (TFP) consists in building a fibrous reinforcement by continuously laying down a tow on a plane backing material which can be a polymer film or a fibrous reinforcement such as woven fabrics or NCF. The tows are maintained in place with the desired orientation with a stitch. To fully take advantage of the design freedom offered by TFP in composite stamping, the designer who can compute the optimized fiber orientations in the final 3D part, has still to determine the tow’s path on the plane backing material since tow’s motions will inevitably occur during the forming process. The present study which falls within the simulation of material forming attempts to model the behavior of TFP preforms using the finite elements method. The backing material is represented by a mesh of 2D elements with embedded linear beam element modeling explicitly the tow’s layout analogously to the modeling of reinforced concrete in civil engineering applications. Consequently the degrees of freedom of the beams are interpolated from the host element. However the displacement constraint at the beam nodes is relaxed in the tow’s path direction. To demonstrate the accuracy of this finite element formulation, standard tests for fiber-reinforced composites such as bias extension test as well as a simple 3D test case will be presented.

During compression of saturated fibrous reinforcement, the flow of the liquid can, under certain conditions, generate fiber-tows washout (in-plane movement of fiber-tows) due to the strong hydrodynamically induced force on the fibrous reinforcement. The objective of the study is first to detect experimentally the onset and measure the full-field magnitude of the flow-induced in-plane deformation occurring during the compression. Secondly, a methodology is presented, aiming at defining the mechanical signature of fiber-tow washout on transverse compression tests. For this purpose, a large-scale semi-transparent test bench has been developed in order to dynamically track the in-plane motion of the fiber-tows occurring in saturated compression. Then, a nomogram is given to identify a process window of a saturated fibrous reinforcement processed by consolidation.

Composites manufacturing using the infusion process involves an unloading phenomenon due to the vacuum bag flexibility. Additionally, during impregnation, woven or non-crimp fabrics exhibit a dual-scale flow. Usual modeling of both phenomena assumes that the fibrous preform is a continuous medium with a varying permeability. Nonetheless, the permeability is affected by the meso-pores size and spatial distribution, which depend on the compaction state. This paper proposes an experimental method to quantify the evolution of a given dual-scale fibrous microstructure under several controlled flow-induced compaction states. A downsized setup has been designed to conduct in situ infusion of quasi-unidirectional fabrics inside a X-ray Computed Tomography device. The downsized setup reproduces large scale infusion phenomena. Using image processing tools, it is observed that, while the stack thickness increases between dry and saturated states, tows swell and displace. The impact of this microstructural reorganization on the in-plane permeability is also identified.

International audience

In this article, we provide a general description of a random stack of solid particles. We consider fine particles of any shape distributed evenly on a flat surface. The angular distribution of the particles is arbitrary. In the absence of constraints imposed on the medium, the volume fraction is expressed analytically according to the parameters of the problem. We thus define a generalized strain measurement of the medium taking into account the internal reorganizations of the microstructure which can then be used in analytical mechanical models to predict their response. The approach is verified digitally by generating digital stacks of fibers, disks and rectangular particles.

In Liquid Composite Molding (LCM) processes, fabric layers are stacked in a mold which may be a few meters long and wide to build up a thickness of not more than a few millimeters. Resin is introduced to fill all the empty spaces between the fibers. As the in plane dimensions are a few orders of magnitude larger than through the thickness, flow of resin through the preform can be modeled using the two-dimensional Darcy’s law, neglecting the through-thickness velocity and assigning the preform an arithmetic averaged permeability from the layers. However, there are situations in which the through-thickness flow is significant where this assumption is no longer valid or justified. To address such cases, a modified averaging scheme was proposed by Calado and Advani (1996) to account for the transverse flow between adjacent layers of a preform and consequently derive an homogenized one-dimensional value of effective permeability. In the current work, such a model is extended to account for the effect of anisotropic off-axis layers in the stack. The result is a generalized scheme for effective permeability averaging layers of heterogeneous preforms, capturing both through-thickness and in-plane effects into a one-dimensional permeability value. This methodology was validated and a parametric study was conducted with different combinations of in-plane and through-thickness permeability values to identify the influence of preform in-plane dimensions and thickness and to define a criteria that relates the material and geometric parameters to the transverse flow.

Fibrous fabrics are used in a variety of applications, among them, for structural composites. Most fabrics are efficiently manufactured from tows or yarns. During textile manufacturing or during fiber reinforced composites manufacturing, the fabrics and tows undergo several movements and deformations. Although there have been several attempts by different authors to model micro structural mechanical behavior of fabrics, they often suffer from unknown geometric dimensions at various loads or unknown materials' mechanical parameters. This paper presents a method for measuring and comparing tows (or yarns) geometrical evolution during compaction perpendicular to the fibers. The tows are compacted and dimensions are continuously measured using confocal chromatograph. Fabrics and tows of the study are composed of glass or carbon fibers, in dry or lubricated states. Volume, compacity evolutions and Poisson's ratio are extracted for a wide range of compaction levels. Tables of material characteristics and experimental data are also provided for a further use or analysis.

The aim of this work is to investigate the viscoelastic effect during acoustoelastic testing of a thermoplastic (polyethylene) for a reliable stress evaluation, which is critical for the processing and in-service life of thermoplastic products. The strain accumulated under step loading-holding stages was firstly characterized by a stepped isostress method, which can be integrated into the acoustoelastic equations. A nonlinear acoustoelastic term was also introduced in order to explain the viscoelastic effect from the increasing stress/strain. Although different holding times and loading rates will change the ultrasonic wave velocity profiles, their influences can be correlated and included in the improved acoustoelastic equations. Furthermore, the nonlinear problem can be circumvented by using an acoustic birefringence technique. Finally, the traditional and nonlinear acoustoelastic coefficients of polyethylene were calculated and verified.

In this study, a simplified cost effective simulation-based methodology is proposed to assist manufacturing engineers in the design and development phase of the resin transfer molding process. Race-tracking is unavoidable in the resin transfer molding and can lead to entrapment of air pockets, which results in parts being discarded as scrap. A purely numerical methodology is presented to distinguish between the critical and non-critical race-tracking scenarios, that will guide the design and production engineers plan an efficient and effective manufacturing strategy. The detection methodology is based on computing the pressure evolution with time during the injection process. The novelty relies on the superimposition of the computed pressure gradient maps that reveals unsuspected common features in the numerous race-tracking cases investigated in the various geometries of increasing complexity considered. The pressure sensors are meant to detect and evaluate different race-tracking scenarios and their level of criticality. The minimum number and locations of pressure sensors arise directly from the highest pressure gradient zones for simple geometries. Sensors placement guidelines are introduced for a simple rectangular shape, then this information is used to qualitatively apply the guidelines to parts of more complex shapes. A general rule that all parts, no matter how complex, can be considered as a combination of simpler ones is presented. Furthermore, a new failure criterion is proposed based on the flow patterns that highlights the likely flow patterns to entrap voids.

Characterizing the response of compressed dry fibrous reinforcements is a key feature to control liquid composite molding processes. Considering preforms manufactured by automated dry fiber placement, supplementary information are needed, since cyclic loading occurs, during which the mechanical properties of tows evolve. In this way, an extensive characterization procedure is proposed in this article. Each mechanical feature, such as nonlinearity, springback, irreversibility, stabilization, and strain rate dependence, is analyzed, and when possible, related to microstructural observations. Particular attention is paid to the response in other directions than the compressed one. Finally, from this characterization, a route to modeling is drawn: analogies with other classes of materials are proposed, leading to choices on the modeling of each mechanical property.

In the frame of LCM processes, the mesoscopic architecture of a preform is defined by the sequential stacking of layers in the mold. Once the compaction is carried out, the precise orientation of layers and spatial distribution of tows is difficult to know with precision. The lack of this information can lead to a wrong prediction of properties. In this work, we present a new non-destructive methodology to obtain information about the mesoscale architecture of a multi-layer preform in its compacted state. Using a flexible pressure-sensitive film, the compaction pressure field is captured. Data is analyzed using spectral Moire theory to retrieve the orientation and spatial distribution of tows for each layer in the stack. This technique is non-destructive and scalable in size, making it suitable to characterize large preforms before infusion. The methodology is detailed and tested over a varying number of layers and configurations.

Fibrous composite reinforcements with enhanced permeability are of a particular interest for liquid composite molding processes requiring the fibrous preforms to be well impregnated by a viscous polymer. The aim of this work is to study the link between the reinforcement permeability and the geometrical parameters of its architecture, taking into account the internal multi-scale porosity distribution by employing the Brinkman equation. In order to design a reinforcement with the enhanced permeability without degrading mechanical properties of a final composite part, the study is conducted with the condition of fixed and high fiber volume fraction. The Proper Generalized Decomposition method, due to its principle of separation of variables, allowed to efficiently compute the solution of the problem for a range of geometrical parameters at once, as opposed to the classical parametric study. A scale separation criterion for in-plane flow was proposed. It estimates when the microscopic intra-yarn flow can be neglected compared to the mesoscopic inter-yarn flow. The major contributing parameters to the enhancement of both the in-plane and through-thickness permeabilities were identified. Principles established in this study were applied to the design of quasi-unidirectional non-crimp fabrics, where the permeability enhancement was evidenced. Keywords: Enhanced permeability, Design, Fibrous reinforcement, Multi-scale structure, Brinkman equation, Proper Generalized Decomposition

Increased application of carbon fiber reinforced polymer (CFRP) composites to resist blast loadings necessitated investigating their dynamic behavior. It requires characterization of the CFRP laminates at high strain rates using split Hopkinson pressure bar or comparable device. In this investigation, tensile properties of the CFRP laminates subjected to high strain rates are studied experimentally and numerically. Experimental evaluation of tensile properties of the CFRP laminates using the Hopkinson bar is a tedious, time consuming, and costly exercise. Hence, upon conducting limited experimental evaluation, a numerical model is developed to characterize these composites at high strain rates. The developed numerical model of the Hopkinson bar technique is validated against the results obtained from the experiments conducted using crossbow system on the CFRP laminates. Full-scale tensile stress–strain curves are developed to study the effect of the strain rate on the ultimate tensile strength, elastic modulus, and energy absorption capacity of the CFRP laminates. It is observed that the CFRP laminates are sensitive to the strain rates and their ultimate tensile strength and energy absorption increases with the increase in the strain rate. Moreover, the effect of variation in thickness and length to width ratio for a constant stress pulse is insignificant.

Ensuring a good inter-laminar bonding while accelerating the production in the tape winding process of thermoplastic tapes is a key question. Finite element models and numerical simulation help to predict the influence of process parameters on the product quality. This article addresses the finite element modeling of the thermal and mechanical phenomena involved in the tape winding process of thermoplastic tapes. It highlights through simulation and experimental validations the influence and necessity to promote the friction between the different components on the product quality and the process itself. This parameter is neglected in the literature which focuses mainly on thermal parameters and occasionally on the mechanical parameters (roller deformations, etc.), excluding the friction.

The fundamental macroscopic material property needed to quantify the flow in a fibrous medium viewed as a porous medium is the permeability. Composite processing models require the permeability as input data to predict flow patterns and pressure fields. As permeability reflects both the magnitude and anisotropy of the fluid/fiber resistance, efficient numerical techniques are needed to solve linear and nonlinear homogenization problems online during the flow simulation. In a previous work the expressions of macroscopic permeability were derived in a double-scale porosity medium for both Newtonian and rheo-thinning resins. In the linear case only a microscopic calculation on a representative volume is required, implying as many microscopic calculations as representative microscopic volumes exist in the whole fibrous structure. In the non-linear case, and even when the porous microstructure can be described by a unique representative volume, microscopic calculation must be carried out many times because the microscale resin viscosity depends on the macroscopic velocity, which in turn depends on the permeability that results from a microscopic calculation. Thus, a nonlinear multi-scale problem results. In this paper an original and efficient offline-online procedure is proposed for the efficient solution of nonlinear flow problems in porous media.

Most Out-of-Autoclave prepregs (OoA) are only partially impregnated with resin. Their impregnation completes during the cure cycle, solely driven by the difference between atmospheric and vacuum pressure. Increased part length leads to an impregnation time gradient caused by the transient air flow inside the fibrous medium. In this work, a novel numerical approach capable of predicting the local impregnation time of a fibrous domain with resin, at isothermal conditions, under the influence of transient air flow, is proposed (delayed air evacuation). Sensitivity studies prove the robustness of the numerical scheme, for a large range of flow time-scales. The same approach is used to predict the local impregnation time of a commercial OoA prepreg tow, for a wide range of part lengths. It is demonstrated that for manufacturing long parts OoA, accurately capturing the influence of the air pressure on the local impregnation state of the tow, is important for quantifying the risk for residual tow porosity.

Continuous fiber-reinforced thermoplastic composites can be produced by compression or consolidation processes at a temperature above the thermoplastic melting temperature. High production rates or high fluidity thermoplastic (TP-HF) viscosities can lead to large in-plane displacements of the fibrous network during the process. The same mechanisms appear when viscous toughened thermosets resins are used. One can assume that the in-plane displacements occur when the liquid thermoplastic flow sets the deformable fibrous reinforcement in motion. The composite material being manufactured is therefore subjected to a hydro-mechanical coupling between a liquid flow and a deformable continuous fibre reinforcement. Within this context, the in-plane flow-induced deformations during transverse consolidation are investigated in this study. An experimental setup is used in order to localize and quantify fibre tow displacements and large strains as a transient full-field measurement. Then, in order to identify the driving forces occurring during consolidation, these induced deformations are taken into account in the modeling. The fibrous reinforcement properties are redefined locally based on the measured full-field strains. The comparison of experimental and modeling results shows that the local in-plane drag force peaks mainly drive the onset of the fibrous architecture deformation.

Consolidation is a crucial step in manufacturing of composite parts with prepregs because its role is to eliminate inter- and intra-ply gaps and porosity. Some thermoset prepreg systems are toughened with thermoplastic particles. Depending on their size, thermoplastic particles can be either located in between plies or distributed within the inter-fibre regions. When subjected to transverse compaction, resin will bleed out of low-viscosity unidirectional prepregs along the fibre direction, whereas one would expect transverse squeeze flow to dominate for higher viscosity prepregs. Recent experimental work showed that the consolidation of uncured toughened prepregs involves complex flow and deformation mechanisms where both bleeding and squeeze flow patterns are observed [1]. Micrographs of compacted and cured samples confirm these features as shown in Fig.1. A phenomenological model was proposed [2] where bleeding flow and squeeze flow are combined. A criterion for the transition from shear flow to resin bleeding was also proposed. However, the micrographs also reveal a resin rich layer between plies which may be contributing to the complex flow mechanisms during the consolidation process. In an effort to provide additional insight into these complex mechanisms, this work focuses on the 3D numerical modelling of the compaction of uncured toughened prepregs in the cross-ply configuration described in [1]. A transversely isotropic fluid model is used to describe the flow behaviour of the plies coupled with interplay resin flow of an isotropic fluid. The multi-scale flow model used is based on [3, 4]. A numerical parametric study is carried out where the resin viscosity, permeability and inter-ply thickness are varied to identify the role of important variables. The squeezing flow and the bleeding flow are compared for a range of process parameters to investigate the coupling and competition between the two flow mechanisms. Figure 4 shows the predicted displacement of the sample edge with the multi-scale compaction model after one time step [3]. The ply distortion and resin flow observed in Fig.1 is qualitatively retrieved by the computational model.

This paper proposes to address the issue of complexity reduction for the numerical simulation of multiscale media in a quasi-periodic setting. We consider a stationary elliptic diffusion equation defined on a domain $D$ such that $\overline{D}$ is the union of cells $\{\overline{D_i}\}_{i\in I}$ and we introduce a two-scale representation by identifying any function $v(x)$ defined on $D$ with a bi-variate function $v(i,y)$, where $i \in I$ relates to the index of the cell containing the point $x$ and $y \in Y$ relates to a local coordinate in a reference cell $Y$. We introduce a weak formulation of the problem in a broken Sobolev space $V(D)$ using a discontinuous Galerkin framework. The problem is then interpreted as a tensor-structured equation by identifying $V(D)$ with a tensor product space $\mathbb{R}^I \otimes V(Y)$ of functions defined over the product set $I\times Y$. Tensor numerical methods are then used in order to exploit approximability properties of quasi-periodic solutions by low-rank tensors., Comment: Changed the choice of test spaces V(D) and X (with regard to regularity) and the argumentation thereof. Corrected proof of proposition 3. Corrected wrong multiplicative factor in proposition 4 and its proof (was 2 instead of 1). Added remark 6 at the end of section 2. Extended remark 7. Added references. Some minor improvements (typos, typesetting)

In the Vacuum Assisted Resin Transfer Molding (VARTM) process to manufacture composites, woven or stitched fabrics are stacked on top of a tool surface and resin is introduced into this porous network by drawing a vacuum. For large parts, to reduce the time for filling, highly permeable distribution media (DM) is placed on top of the fabric layers to accelerate the in-plane filling process. Many factors such as the manufacturing process, handling, variation in fabric manufacturing and placement of fabric cause heterogeneity in the permeability of fibrous materials. Due to the presence of the DM, the heterogeneous through the thickness permeability (Kpin) of a fabric can dramatically affect the flow of resin and cause air pockets or voids which are mechanical flaws resulting in the rejection of the composite as scrap. Statistical characterization of Kpin is crucial for understanding the (i) effect of heterogeneity in Kpin and its interaction with DM permeability and void formation and (ii) for generating the field of random numbers (Kpin), which can be used for simulations to predict resin flow and void formation for such materials that exhibit stochastic variability. The novelty of this study is that the observed random field (Kpin) is generated for numerical simulation through statistical analysis. First, in this study, the heterogeneity in Kpin was statistically characterized by spatial correlation with Moran's I index and semi-variogram. Then the random field of Kpin was generated by transforming the normal numbers from Karhunen–Loève (KL) expansion to gamma numbers. A numerical flow simulation of the VARTM process with the generated random fields was performed using Monte Carlo method for three types of Distribution Media (DM). The outcome is compared with experimental results and to simulation results that used experimentally determined Kpin data as an input. © 2018 Elsevier Ltd

The construction of mesoscopic models of fiber washing effects during the high pressure resin transfer molding (RTM) process with thermoplastic polymers requires the knowledge of the limit of adhesion coefficients between dry and impregnated fabric reinforcement plies and the mold. To measure these coefficients, a new method is described. During RTM process, a preform is placed in the cavity. The mold is closed to reach a given cavity thickness. Therefore, the normal stress applied to the preform can be relaxed before injection. In order to comply with these RTM specificities, the limit of adhesion coefficients are measured on relaxed preforms at given thicknesses. Hence, a first set of compression tests followed by a relaxation step are performed in order to measure the normal stresses the preforms undergo. The relaxation can reduce compression normal stress by 30%. The relaxation is taken into account for the limit of adhesion calculation. Then a second setup has been built in order to measure forces du...

The acoustic birefringence method has been used since a long time for metallic materials to detect residual stresses. This method is rarely used for polymers and composites due to the strong attenuation of the acoustic signal. Moreover, for thermoplastic polymer, with a slight anisotropy (known as texture for metals), coupling with applied stresses, extends the complexity of the residual stresses determination. The slight anisotropy of thermoplastic polymer should be studied with shear waves prior to the residual stress determination. If the material is perfectly isotropic, we are able to detect and quantify the residual stresses. If the material is slightly anisotropic, it is also possible to evaluate residual stresses by taking into account the directions of the principal axes of material anisotropy. POLYM. ENG. SCI., 55:2307–2312, 2015. © 2015 Society of Plastics Engineers

A crucial step in many composites manufacturing processes is the impregnation of fibrous medium with the resin. The fundamental property needed to quantify the flow is the permeability of the fibrous medium. Process models require the permeability as input data to predict flow patterns and pressure fields. Image-based computation can offer a good alternative to provide input data for subsequent analysis. However, digital images contain a huge amount of information that is difficult to handle in numerical models. Then efficient numerical techniques are needed to solve homogenization problems with geometrical data coming from high-resolution images, involving two or three scales and linear and non-linear fluids. Within this framework, this work addresses three main questions: (i) how to define an equivalent macroscopic Darcy’s model from a microscopic description consisting of a viscous fluid flow model defined in a two and three scales porous medium?; (ii) the discussion on the existence of an intrinsic geometrical permeability tensor in the general case of non Newtonian rheothining fluid models, and (iii) the proposal of a constructive multi-scale strategy for performing micro-macro simulations in both the linear and the nonlinear case.

Image-based simulation is being considered more and more in industrial applications, where more than simulate the expected reality it seems preferable to simulate the reality itself. For example, in heterogeneous composite materials more than assuming a particular distribution of the material constituents, it is preferable to solve the model in the real distribution, the one existing in the considered material that is accessible from an appropriate scanning technique. However such route requires, in many applications, advanced simulation techniques able to proceed in real time, looking for a synchronism with the scanning process itself. This constitutes a challenging issue due to the rate of image acquisition and the amount of information associated to each one. This work proposes an efficient strategy for solving thermal problems in two scales heterogeneous media in both the linear and the nonlinear case. In both cases the calculation of the homogenized thermal conductivity tensor requires the solution of many thermal problems defined at the microstructure level, here considered coming from digital images. For alleviating the computational cost and speeding up calculations to fulfill real time constraints, we propose an offline/online procedure for evaluating the homogenized thermal conductivities.

In this study the multi-objective optimization of a fiber reinforced composite part will be presented. Additionally to the weight and the mechanical properties, its manufacturability with the Resin Transfer Molding process is considered during the optimization. For the last mentioned, a new CPU-time efficient method, called Process Estimator, was developed. Based on the local material parameters it allows to calculate the filling times of complex and fine meshed parts with varying permeabilities and porosities in some seconds to minutes. The rapidity of the method will be illustrated with the help of three example parts. These examples demonstrate the importance of the process consideration during a mechanical design optimization.

The aim of this work is to investigate the acoustoelastic effect in a thermoplastic polymer (polyamide 6) under uniaxial compression using ultrasonic waves (longitudinal and shear). The acoustic velocities varied linearly, as expected, when small strains or stresses were applied, from which Murnaghan coefficients were obtained for the first time for this type of polymer. This relationship changed to be nonlinear above a critical strain at about 0.7%. This nonlinear behaviour has already been observed for polymers under tensile tests but without interpretation. In order to explain the nonlinear part of the curves, an analysis was conducted from a thermodynamic standpoint and ultrasonic measurements were related to the molecular structure of the polymer. The Gruneisen parameter was introduced as an index and its magnitude showed how the system changes from anharmonicity to harmonicity of interatomic (intermolecular and intramolecular) interactions according to the applied strain.

International audience; Liquid Composite Molding (LCM) processes are increasingly used to produce composite parts. Most of those processes involve compression of the fiber reinforcement and resin flow. In order to accurately model LCM processes, a good knowledge of fiber reinforcement behavior in compression is required. Several models have already been published, but none of them include permanent deformations. Also because of the large deformation involved in the processes, a finite strain formulation is proposed. Results are given for a glass twill‐weave fabric.

International audience; Composites manufacturing processes usually involve multiscale models in both space and time, highly non-linear and anisotropic behaviors, strongly coupled multiphysics and complex geometries. In this framework, the use of simulation for real-time decision making directly in the manufacturing facility is still precluded nowadays, in spite of the impressive progresses reached in numerical analysis and computer science during the last decade. In this paper, a process-specific simulation tool based on reduced order modeling is introduced, the Simulation App. This concept is presented through a practical case involving a multi-physics and coupled problem describing the manufacturing process of a composite outlet guide vane. We show that several manufacturing settings can be simulated in few seconds with the Simulation App, thus enabling fast process optimization. Finally, the advantages over general-purpose simulation software, in the context of process simulation, are discussed.

International audience; Several studies showed that thin-plies composite laminates apparently exhibit higher strength than its counterpart using thicker plies. In this work, a high-resolution numerical discretization based on the use of in-plane-out-of-plane separated representations within the Proper Generalized Decomposition PGD framework is performed in different composites laminates for determining the effect of ply-thickness on the laminate stiffness. The results reveal that the global stiffness increases with the ply-thickness reduction. The analysis also reveals that this mechanical enhancement is due to structural effects and not to the mechanical confinement initially advanced.

Bottom ash resulting from the incineration of various domestic wastes can be viewed as a typical granular material. It is mainly used in civil engineering as a substitute for traditional natural aggregates. The purpose of this paper is to characterize their mechanical behavior and evaluate their mechanical properties for engineering applications. First, results of triaxial tests confirm that bottom ash behaves like dense sand. Second, the deformation and strength characteristics of bottom ash, such as the secant modulus, Poisson ratio, characteristic angle, dilation angle, effective cohesion and effective friction angle, are determined. It is found that these mechanical parameters are in close agreement with those of road aggregates and are influenced by the effective confining pressure. Third, the evolution of the deformation modulus according to the axial strain and the variation of the deviator stress according to the mean effective pressure are analyzed. Finally, a set of points of the yielding state is determined from triaxial tests to represent the shape of the yielding surface of bottom ash. © 2017

A new dynamic method based on bilayer system is proposed to characterize the residual stress formation during the crystallization of semi-crystalline polyethylene glycol 10000 (PEG10000). The resin is coated on a solid polymeric film to create a filmsubstrate compound. Its temperature field and dynamic deflection are monitored by synchronized optical and thermography cameras. The crystallization kinetics is first characterized from the former information. Then a simple dynamic model is proposed to relate the dynamic deflection with crystallization process. Residual stresses are established and in the range of 0–2.1 MPa. The generation of residual stresses is due to the edge constraints of the cantilever beam and to the increase of viscosity during solidification that allows the polymer to carry tensile loadings. The spherulite impingement is found to be important for this period from a microscopic view. Boundary condition should be well controlled to steer residual stresses. Such method is promising to measure residual stresses at the micro-scale for polymers to be spread on a flexible substrate and can mimic different mechanical situations of interest.

International audience; The superior material properties of carbon fiber-reinforced composites make them especially attractive for applications in aeronautics and aerospace industries. Cost reduction and time saving are continuously driving industry, leading to new industrial challenges which include manufacturing composite structures with optimal mechanical performances using the potential of advanced processes using robotics.To produce complex part shapes, technologies implying fabric draping in a mold imply large waste amount, fabric structure variability and uncertainties concerning local fiber volume fraction amount and thus final mechanical properties. To overcome such issues and comply with cost and time efficiency, automated dry fiber placement for preform manufacturing is proposed. This approach allows to integrate many functions in a complex part thank to the ability of the robot to steer fiber tows at specific locations. The final composite part is obtained by injecting the produced preform with resin using RTM (Resin Transfer Molding) or infusion process.The presented project aims to define the influence of the process driving parameters during fiber placement on the final preform properties range. Preforms were produced using a lab-scale automated placement demonstrator. Three preforms configurations were tested to highlight the influence of the preform structure on permeability and mechanical parameters through characterization of the compression behavior and permeability of the produced preforms. Choice of configuration will affect mechanical properties on the manufactured preforms, whereas creation of open channels to enhance the flow propagation during manufacturing does not necessarily increase the preform permeability.