Composite Damage Model Research Articles

Infant skull fractures align with the direction of bone mineralization.

The geometry and mechanical properties of infant skull bones differ significantly from those of adults. Over the past decades, debates surrounding whether fractures in infants come from deliberate abuse or accidents have generated significant impacts in both legal and societal contexts. However, the etiology of infant skull fractures remains unclear, which motivates this study with two main components of work. Firstly, we present and implement a progressive unidirectional fabric composite damage model for infant cranial vaults to represent ductile and anisotropic properties-two typical mechanical characteristics of infant skulls. Secondly, we hypothesize that these intrinsic material properties cause injuries perpendicular to the fiber direction to dominate infant skull fractures, resulting in fracture lines that align with the direction of mineralization in the infant skull. The material model and the finite element (FE) model were verified hierarchically, and this hypothesis was verified by reconstructing two legal cases with known fall heights and implementing the above damage model into CT-based subject-specific infant FE head models. We discovered that the infant skull is more susceptible to injuries within planes perpendicular to the mineralization direction because of the anisotropic mechanical property caused by the direction of mineralization, leading to infant skull fractures aligning with the mineralization direction. Our findings corroborated the several previously reported observations of fractures on cranial vaults, demonstrating that these fractures were closely associated with sutures and oriented along the mineralization direction, and revealed the underlying mechanisms of infant skull fracture pattern. The modeling methods and results of this study will serve as an anchor point for more rigorous investigations of infant skull fractures, ultimately aiming to provide convincing biomechanical evidence to aid forensic diagnoses of abusive head trauma.

Scaler nonlinear continuum damage models for ceramic matrix composites with and without in plane ply anisotropy at room temperature

Damage tolerant design of aerospace structures with continuous fiber reinforced ceramic matrix composites (CMCs) require progressive damage models that can predict the behavior at local stress concentrations. Here we introduce three versions of a continuum damage model including a two-variable damage model that considers damage formulation perpendicular and in line with the direction of the primary fiber reinforcement, a coupled two-variable damage model and a four-variable damage model that includes the previously mentioned modes along with two additional modes specific to shear damage. The model is demonstrated using data for two different SiC/SiC CMCs including one that consists of stacks of balanced woven plies and one that consists of stacks of unidirectional laminate-based plies. The behavior is well captured for the balanced woven ply-based materials, but required a more complex damage formulation to close the solution for the unidirectional based plies that exhibited a significant degree of anisotropy.

A Review of Machine Learning for Progressive Damage Modelling of Fiber-Reinforced Composites

A Review of Machine Learning for Progressive Damage Modelling of Fiber-Reinforced Composites

Numerical study of high-velocity impact on aeronautic carbon composites

This paper presents research concerning the numerical analysis of two new carbon composite materials subjected to high-velocity impact. These new composites were produced using new out-of-autoclave processes, making them more environmentally friendly. An experimental campaign was conducted on these materials in order to determine the effects of a metal sphere during impact. A numerical model using ABAQUS explicit is assembled using shell finite element models, and a non-linear dynamic analysis is applied. For the constitutive relation of the material, a fiber-reinforced composite damage model from the ABAQUS library is used. This initial numerical model is then compared with the outputs of the experimental campaign in order to validated it. Next, a numerical campaign is performed to understand carbon composites response to the impact of a hailstorm strike, in which it is possible to verify that one of the specimens presented higher performance, by displaying less penetration and damaged area.

Probabilistic progressive damage modeling of hybrid composites

A novel analytical probabilistic progressive damage model (PPDM) is introduced for multiphase composites to predict the damage behavior of hybrid composites. The PPDM is based on effective field methods and the stochastic nature of fiber damage is captured by including weakest link theory and Weibull statistics. Three additional models are developed to compare with the PPDM. A stochastic model analogous to the PPDM (called SPDM), and two finite element models, one stochastic (SFEM) and one probabilistic (PFEM). All models are developed in a thermodynamically consistent framework and are extended to include residual thermal stresses. Finally, the four models are compared with models from the open literature for an AS4-M50S hybrid carbon–carbon composite with different hybridization ratios of high to low elongation fibers. The comparison reveals a great agreement between all models and indicates that the stochastic nature of fiber damage is the most influential parameter leading to damage.

Fatigue damage evolution of modified recycled aggregate concrete

Basalt fiber (BF) and nano-silica (NS) are used to improve the properties of recycled aggregate concrete (RAC) and produce modified RAC (MRAC). The fatigue failure patterns of the MRAC were observed through a uniaxial compression fatigue test, and the reasons for the different patterns were explained. The effects of the stress level, replacement rate, and loading frequency on the fatigue life were investigated, and the effects of BF and NS on the fatigue strain, damage evolution, and energy dissipation evolution of MRAC were analyzed. The enhancement mechanism of BF and NS on RAC was revealed using microstructural analysis techniques, such as scanning electron microscopy (SEM) and computed tomography (CT). The results showed that the fatigue failure modes of the MRAC could be divided into split, shear, and mixed failures. The fatigue life and fatigue strain of RAC exhibited a pattern of initially increasing and subsequently decreasing with an increase in the BF content (from 0 kg m−3 to 3 kg m−3 and then to 6 kg m−3). At stress level of 0.80, the average fatigue life of the specimens increased by 76.15 % compared to RAC when the BF doping was 3 kg m−3; however, when the BF doping was increased to 6 kg m−3, the average fatigue life of the specimens increased by only 47.68 % compared to RAC. The mixed BF and NS improved the fatigue life of RAC more than the single mixed BF or NS. Fatigue equations and unified single logarithmic fatigue equations, including fiber parameters, were established to accurately describe the relationship between the fatigue stress level, fatigue life, and failure probability. The fatigue strain gradually decreases with increasing NS content (0–2 %); with increasing loading frequency (from 1 Hz to 10 Hz and then to 15 Hz), the fatigue strain gradually decreases. The fatigue life of the MRAC decreased dramatically at low loading frequencies (1 Hz) and then steadily increased as the loading frequency increased. BF and NS have an improving effect on the porosity of microstructure, improve the fatigue life of RAC, and BF exhibits a more significant improvement than NS. A composite fatigue damage model that comprehensively considers the contribution rates of NS and BF was established, and the fatigue damage evolution of the MRAC was predicted.

A new progressive fatigue damage model for the three-dimensional braided composites subjected to locally-variable-amplitude loading

The fatigue-induced stress redistribution in three-dimensional (3D) braided composites, even when subjected to externally constant-amplitude fatigue loads, results in locally-variable-amplitude fatigue loading, significantly impacting their fatigue performance degradation. This paper is aimed to propose a novel fatigue damage analysis method to predict the fatigue behavior of 3D braided composites subjected to locally-variable-amplitude loading. To this end, the continuous damage method, residual strength, residual stiffness, and fatigue life models are combined to characterize the properties degradation of 3D braided composites with cycling. Once stress redistribution occurs, the fatigue properties degrade based on the new stress level, and the equivalent cycle number is recalculated accordingly. As the material undergoes different cycle blocks, damage accumulates, ultimately leading to failure when the overall stiffness of the material degrades to 80%. Traditional fatigue models are also employed to predict the fatigue behavior of representative volume elements for comparison. The results predicted by the present model shows good agreement with experiments. Furthermore, the influence of stress levels and architectures of 3D braided composites on fatigue performance is systematically discussed.

Navigating the unknown: Tackling high-dimensional challenges in composite damage modeling with bootstrapping and Bayesian uncertainty quantification

A methodology for quantifying uncertainty in high-dimensional problems involving sophisticated progressive damage models for composites is proposed. This approach combines bootstrapping and Bayesian Uncertainty Quantification to determine the distribution of Quantities of Interest typically related to structural response. Applied to a practical case involving a 40-dimensional problem with Open Hole Tension and Compression specimens, where stochastic variables include material and geometric parameters, as well as ply orientations, the methodology efficiently provides consistent predictions of QoI distributions. Its effectiveness makes it highly suitable for real industrial applications, enhancing capabilities in the design of composite structures.

Development of a new progressive damage model for woven fabric composites

There is no native support for the modeling of woven fabric composites within Abaqus. A new approach has been implemented within Abaqus, which is shown to be robust and capable of modeling woven composites across a range of loading conditions. This approach predicted the bending modulus and displacement at failure of woven carbon-fiber composites under three-point bending within the standard deviation of experimental results. Under high-velocity impact loading, the critical perforation velocity of woven aramid panels was predicted to within 3% of the experimental results, while the back face deformation to within the experimental error.

A coupled ductile damage model for metal matrix composites: Development and application

The prediction of failure behavior in metal matrix composites remains a significant challenge in both composition design and process optimization. An accurate prediction of metal matrix composites damage evolution is a crucial for enhancing the quality of metal matrix composites forming. As the material undergoes plastic deformation, it experiences void initiation and growth, resulting in consequential microstructural transformations, stiffness degradation, and mechanical property shifts. In this work, we employed a model to predict damage progression and stiffness decay in metal matrix composites. Leveraging the Gurson-Tvergaard-Needleman framework, this homogenization model accounts for the impact of the evolution of voids and reinforcing phases, on the composite's mechanical properties. The influences of reinforcing phases on voids nucleation and growth were particularly considered, and also the interaction of voids, matrix, reinforcing phases, and stiffness were integrated to discuss their impacts on damage evolution and mechanical performances of the metal matrix composites. The model was implemented as an Abaqus VUMAT subroutine, with its validity gauged by analyzing the influence of model parameters on failure mechanisms and inherent elastoplastic traits. Utilizing the flanging process of carbon nanotube-reinforced aluminum matrix composites as a case study, a significant agreement was observed between experimental and simulated force-displacement profiles, as well as crack evolution routes.

A progressive damage model for 3D woven composites integrating kinking-buckling competition

In this work, a progressive damage model (Kinking-Buckling PDM) considering fiber micro-buckling failure, fiber kinking failure, transverse inter-fiber failure of yarns, matrix failure and interface debonding, is proposed to predict the compressive mechanical behavior of 3D woven composites (3DWC). Fiber kinking and fiber micro-buckling failure are adopted in the form of competition to characterize the longitudinal failure of fiber yarns. A bilinear damage evolution model is established to control the damage accumulation of each constituent. Meanwhile, a material tangent constitutive tensor characterizing the constitutive behavior of 3DWC is derived. The complex failure process of 3DWC under compression is displayed through Kinking-Buckling PDM. As the fiber initial misalignment angle increases, the compressive strength of 3DWC declines significantly, meanwhile, the main failure mode of 3DWC will be transformed from fiber micro-buckling to fiber kinking in warp yarns. The relevant experimental results are used to confirm the accuracy of the PDM.

Αn efficient numerical model for the simulation of debonding of adhesively bonded titanium/CFRP samples induced by repeated symmetric laser shocks

ABSTRACT Novel engine fan blades are made of 3D woven composite materials and incorporate a protective metallic layer at the leading edge. The end-of-life of such structures involves complicated disassembly and recycling processes. In this paper, laser-shock is being investigated as an environmentally friendly disassembly method. In this context, symmetric laser-shock experiments that have been conducted in a previous work using a time delay between the shots have been proven successful for debonding initiation and propagation as they manage to effectively focus the tensile stress field developed by the shock waves at the bondline. In this paper, a numerical model simulating the symmetric laser-shock disassembly of titanium/CFRP specimens has been developed using the LS-Dyna explicit FE code. The aim is to give a deeper insight into the physical mechanisms and to optimize the experimental process. To simulate the behavior of the titanium part, the Johnson-Cook material model in combination with Grüneisen equation of state has been employed, while for the composite part a progressive damage material model incorporating high-strain rate effects has been used. The bondline between the two parts has been modeled using cohesive zone elements. To obtain tensile and shear elastic and strength properties at high strain rates for the composite damage model, tension and punch shear Split-Hopkinson Bar tests have been conducted. The numerical results correlate well with back-face velocity profiles obtained from single sided laser shock experiments and with debonding patterns in the adhesive, observed in electron microscopy images, induced by symmetrical laser shock experiments. After the validation, the numerical model has been used successfully to simulate a larger disassembly process composed of 16 consecutive shots.

An improved composite impact damage model and bird-striking damage analysis with smoothed particle hydrodynamics-finite element method

An improved impact damage model is proposed, which can be used to analyze the progressive failure behavior of fiber-reinforced composites. The model is constructed based on the 3D Hashin failure criterion. However, the difference is that the strain component along the thickness direction in the matrix energy evolution model is introduced, and the effects of and on impact damage are added. Rigid-body and bird-body impact problems are studied, which validates the accuracy of the numerical model by comparison with experiments. A 3D numerical model of light aircraft fuel tank impact is constructed, and the SPH-FEM coupling method is used to predict the bird strike failure problem. The structural damage and impact behavior of a light aircraft fuel tank under impact loading are analyzed. The results show that, under different impact energies, the damage to the matrix is more serious and expands along the fiber direction, and the structure presents different damage forms. This paper provides an effective research model for the study of drop shock and the airworthiness of composite fuel tanks.

Damage mechanisms of composite laminates under impact loading including the effect of pre-load

The focus of the present research is on the damage modelling of composite laminates subjected to Low Velocity Impact (LVI). This research also shows how this model can be employed to assess the effect of pre-load. The composite damage model, integrating a VUMAT subroutine and a cohesive zone model, is first validated against the results obtained from conventional impacts on laminates without pre-load. The validated model is then employed to predict the impact response of pre-loaded composite laminates. It was found that the pre-load influences the load response, damage behaviour and energy absorption mechanisms of composite laminates. This is with pre-tension increases the maximum load during the impact event, whereas pre-compression has the opposite effect. Matrix cracking and delamination are found to be reduced slightly due to pre-tension, whilst matrix cracking and delamination are considerably increased by pre-compression. The developed model can provide guidance to the design and maintenance of composite components subjected to pre-load and impact.

Multiscale progressive damage model for plain woven composites

In this paper, a novel multiscale progressive damage model is proposed to study the complex mechanical behavior and damage mechanism of plain woven composites. In the macroscale, composites are treated as full-scale model with three-dimensional solid elements. The mechanical response of element integration points depends on the mesoscale constitutive relationship, including orthotropic undulating yarn and isotropic matrix. Based on the microscopic representative volume element (RVE), the stress amplification factor (SAF) database is introduced to characterize the microscale stress state of the fiber and matrix. SAF is used to realize the stress correlation calculation from the yarn level to the micro level, and a progressive damage model based on the micro-mechanics of failure (MMF) theory is developed. Quasi-static tension experiments are carried out to validate the accuracy of the model, and the influence of meso‑scale waviness and non-waviness constitutive models on the simulation results is compared. The results indicate the waviness constitutive model predicts are more consistent with the experimental results, and the prediction deviations about modulus, failure strength and failure strain are around 5%. The damage initiation of matrix in the yarn and pure matrix occurs successively, and finally the fiber failure is the main cause of catastrophic damage. Furthermore, the proposed multiscale method can be further used to predict the strength and damage of full-size woven composites under complicated loading.

Nonlinear progressive damage model for woven patch-repaired laminate composites

Laminated fiber-reinforced composite materials are vulnerable to damages arising due to impact events like accidental tool drops, bird strikes, etc., which significantly reduce the structural integrity of these materials. Therefore, one of the popular adhesively bonded repair techniques, i.e. patch repair is employed in this work. The present work focuses on developing a three-dimensional energy-based gradual degradation model for patch-repaired laminates, including the effects of shear non-linearity under tensile load. The model utilizes modified Hashin failure criteria for plain woven laminate and maximum shear stress criteria for the adhesive. Using the nonlinear model, the failure strengths for different pristine, drilled and repaired specimens with double-sided patches of [0]2 and [45]2 are studied. Tensile experiments in accordance with ASTM-D3039/D3039M are performed using Digital Image Correlation (DIC) and Acoustic Emission (AE) to validate numerical results. The simulated results based on the proposed model and experimental observations are used to compare the damage mechanisms of pristine, drilled, and patch-repaired laminates. It is observed that matrix cracking is majorly responsible for damage initiation. Subsequently, fiber–matrix debonding and fiber breakage account for the ultimate failure of pristine, drilled and repaired specimens. The developed model is used for studying the effect of different patch sizes of repaired specimens under tensile load.

An explicit multi-phase field damage model for long fiber-reinforced composites

An explicit multi-phase field damage model for long fiber-reinforced composites

A general continuum damage model for soft composites

Soft composites, including filled rubbers, nanocomposite gels, double network hydrogels, and multiple network elastomers, all exhibit stress-softening behavior in loading–unloading cycles, also denoted as the Mullins effect. Though qualitatively this damage behavior shows similar features, the details still vary with material types. In this work, we provide a general approach towards modeling the Mullins effect for soft composites based on continuum damage mechanics. We start with formulating a one-dimensional damage model. A logarithmic damage variable is employed, while a damage potential function is adopted to obtain the evolution law for the damage variable. We then extend the one-dimensional damage model to three-dimensions. The eight-chain type micro–macro transition is used to formulate an isotropic damage model. An anisotropic damage model is constructed based on the full network model, while a new micro–macro mapping is used to determine the micro stretch from the macro deformation. With this new mapping, the full network model appears in a consistent format in that the free energy, dissipation, and stress are respectively given by the average of their directional micro counterparts. In the current approach, two constitutive functions need to be determined: the elastic free energy density and the stored energy limit. Simple forms of these two constitutive functions are provided with only a small set of model parameters. The resulting models are then successfully applied to describe the stress response in loading–unloading cycles and the anisotropic response of predeformed materials for different kinds of soft composites.

On a computational stress-based non-local damage model for quasi-brittle composites

Numerical models for the stress and strain analysis of quasi-brittle composites, such as cement-based ones, supplied by various stiffening particles, under mechanical, thermal, etc. loads should handle i) the creation of certain micro-damaged zones, antecedent to ii) the initiation and propagation of a system of macroscopic cracks. Whereas ii) can be analysed using some extended, generalized or similar finite element technique (XFEM, GFEM, etc. algorithms), i) must rely on a smeared crack formulation, whose regularization properties are derived from a non-local stress evaluation. This paper studies mathematical and computational properties of such model, based on the Eringen's approach, useful in many applications in civil engineering.

A Multiaxial Fatigue Damage Model Based on Constant Life Diagrams for Polymer Fiber-Reinforced Laminates.

In the last decade, fatigue damage models for fiber-reinforced polymer composites have been developed assuming the fracture energy equivalence hypothesis. These models are able to predict a fatigue life of composite laminates, but their identification requires a significant number of off-axial tests for various stress ratios. The present study proposes the stress ratio dependent model, which phenomenologically adopts a decrease in stiffness and residual strength of a unique ply according to experimental constant life diagrams. Hashin, Tsai-Hill, and the maximum stress failure criteria are utilized for damage initiation considering the residual strength of the ply. The obtained results indicate a sufficiency of using S-N curves for UD 0°, UD 45°, and UD 90° for identification of the model. The model was verified by S-N curves for UD 10°, UD 15°, and UD 30° and its applicability was demonstrated for prediction of a fatigue life of composite laminates with an arbitrary lay-up. The model is implemented into ABAQUS finite element software as a user subroutine.