Damage modelling strategies for unidirectional laminates subjected to impact using CZM and orthotropic plasticity law
Introduction
The design of composite structures typically follows the Building Block Approach (BBA) [1], suitable for investigating the components at various levels of complexity. Using FE models, the above method aims to mimic the mechanical behaviour of complex structures starting from coupons under simple loading scenarios, and moving through tests of sub-components and components, up to the full-scale mechanical system [2], [3]. Since, the BBA is costly and time-consuming process the minimization of the experimental tests is pursued as well as the development of FE models has to comply with the accuracy of the failure loads and with a moderate computational effort.
In the framework of BBA, the experimental evaluation of Carbon-Fibre Reinforced Polymer (CFRP) crashworthiness and its prediction using FE analysis are applied. In fact, the impact tests are investigated at low- [4], high- [5], and ballistic velocity [6]. The LVI induces a minimal surface damage and internal extended failure mechanisms [7]; this consists of intralaminar damage (e.g. fibre kinking due to compression in the top part of impact sample and matrix cracking due to bending on the bottom ply), and interlaminar delamination among adjacent layers [8].
Analytical studies apply a fracture mechanics approach to predict the delamination threshold forces and damaged area due to impact on laminates [9], [10], [11], [12], [13], [14], [15]. In fact, Davies et al. have demonstrated that the delamination mechanism is driven by the interlaminar sliding shear failure mode (mode II) and by the associated energy release rate (GIIC) [9], [10], [12]; Christoforou have derived the same formula in a normalized form [15], whereas other authors have included the number of delaminated interfaces for the evaluation of the threshold force [11], [13], [14]. The limitation of these methods lies in the hypothesis that the delamination area is equal at each interface among plies, which does not find the experimental counterpart. Nevertheless, in the present work, the analytical model developed in [9] has been adopted for the evaluation of the energy release rate of mode II (GIIC) replacing peculiar experimental test.
Alternatively, FE models are powerful in simulating the impact response of composite laminates; however, their accuracy is affected by the material law as well as by the damage modelling technique [16]. The constitutive material law needs to mimic the plastic field, the failure onset (i.e. the intralaminar and interlaminar failures) and the crack propagation.
Including the plastic field of the material, some studies consider the in-plane-shear stress–strain by fitting the experimental data using a) an exponential law [17], b) a third-grade polynomial [18] or c) a quadratic yield surface [19]. However, the above models assume that the material is linear elastic in compression before damage. Since the laminate plastic response is driven by the matrix properties in compression, this assumption might lead to underestimate the loss energy due to impact, see for instance [20]. Therefore, concerning this aspect, an orthotropic plasticity law has been developed here, to accurately predict the permanent indentation and the energy loss induced by the non-linear response of matrix in compression.
The intralaminar damage onset has been analysed by the implementation of failure criteria. Li et al. compare the force–time and the force–displacement curves due to LVI via FE for the following [20] criteria: maximum stress, Hashin [21], Hou [22], Tsai-Wu [23], and Puck [24]. In [20], it has been demonstrated that the laminate mechanical response is well-predicted by those failure criteria; however, the Tsai-Wu criterion exhibits a higher permanent indentation since it does not consider separated damage modes.
Furthermore, the Continuum Damage Mechanics approach (CDM) is used for the intralaminar damage propagation modelling; in fact, the crack propagation is investigated at the macroscopic level defining peculiar damage variables which reduce the material mechanical properties [25], [26].
Regarding the FE models, the selection of the element spatial scale is crucial since it affects the capability of damage modelling and the computational effort. The FE models are tuned to investigate the material at various scale-levels, ranging from the macro (i.e. layered shell model), to the micro-scale, see Fig. 1 of [26]. A shell-based macro-scale model combines an accurate prediction of the intralaminar failure mechanism [27]. However, the models lack in the prediction of delamination.
Recently, the literature has been endowed with meso-scale models, this approach has been applied for the simulation of the LVI [28], [29], [30], [31], of the compression after impact test [32], [33], and of the repeated impact response [34], and also for the experimental–numerical characterization of hybrid aluminium CFRP laminate [35]. In such models, each ply is modelled by at least one solid element through its thickness, and the interface between the plies is mimicked by CZM [36], [37].
At the micro-scale level, Miao et al. [38] have investigated the impact strength of a 3D fabric structure by FE. These models are used to collect the average behaviour of the material which are adopted on macro-scale models.
On the other hand, accurate models based on global-to-local approach have been developed by [31], [39], [40], [41]. In the above works, the laminate is modelled with a fine, meso-scale solid mesh, located at the impact zone where the damage occurs, while a coarser macro-scale mesh is adopted in those areas of the laminate where damage does not occur.
For instance, the global-to-local approach is applied to simulate automotive and high-value aerospace components during crash events when the impact damage occurs only locally [26]. Unfortunately, an impact involves many components and extended areas, for instance the peculiarities of a car crash event are discussed in [42], [43], [44]. Thus, the adoption of solid elements is minimized to contain the computational effort of these FE analyses. Therefore, for a general approach, the global-to-local strategy is abandoned in the present work.
This paper reports a methodology to simulate damage evolution in composite structures subjected to impact in the early phase of the BBA. An experimental–numerical correlation of a thick UniDirectional (UD) CFRP laminate subjected by a LVI test is pursued.
Firstly, the crashworthiness and the elastic–plastic CFRP properties are detected during a LVI test according to the ASTM D7136 standard [45]. The internal damage is evaluated and discussed by means of non-destructive micro-Computerized Tomography (CT) analysis.
Secondly, two numerical strategies have been developed for the impact prediction of the UD lamina using a) a shell-based model and b) a solid-based model. The composite damage onset is modelled using a modified Tsai-Wu criterion which includes separated failure modes, a CDM approach are employed to simulate the damage propagation mechanism, whereas in the solid model, the delamination is mimicked using CZM.
A non-conventional approach has been proposed for the evaluation of the energy release rate associated with mode II (GIIC) by means of analytical formulation, this model rationalize the delamination propagation of a quasi-isotropic laminate subjected to transverse load. The proposed analytical approach allows to reduce the number of mandatory tests for the BBA approach with respect to a more traditional one.
The FE forecasts are in good agreement with the experimental data in terms of internal damage, force, energy, and displacement versus time. In fact, the error on maximum impact force is 4.0 and −2.3 per cent for shell-based and solid-based model, respectively.
Section snippets
Material characteristics
The composite material is constituted by continuous UD High-Strength T700 carbon reinforcement and epoxy resin. Carbon fibre tows are transversely stitched with polyester yarns to ensure the reinforcement alignment.
The laminates are obtained by stacking UD lamina and consolidated via compression moulding technique. The process is performed at a temperature of 155 degreesC and a pressure of 110 bar with a curing time of 8 min.
The average fibre volume fraction is 55 per cent. The UD lamina has a
Numerical investigation
Two FE models are developed for the prediction of the laminate behaviour due to LVI, namely shell-based and solid-based numerical models.
The software package employed for the numerical simulation is the explicit solver Altair® Radioss [58], while the model set up is made with Altair HyperMesh 2019.1. LAW25 and LAW83 are employed for the composite lamina modelling and for the CZM, respectively.
The material law and the damage parameters are discussed in the following. The composite is modelled
Experimental-numerical comparison
In the following, FE forecasts are compared to the experimental results in terms of: structural response (i.e., force–time, force–displacement, and absorbed energy-time curves), internal damage, and finally, permanent indentation. The comparison is reported in Fig. 21, Fig. 22, Fig. 23, where the experimental data uncertainty is highlighted.
In Fig. 21, the shell-based Model A shows a higher stiffness compared to the Model B and the actual specimen; this is due to shell element formulation. The
Conclusions
A thick CFRP UD lamina has been experimentally and numerically characterized. A LVI is performed on five specimens of a laminate with a quasi‐isotropic stacking sequence, i.e. [-45/90/45/0]s. The impact energy is 36.3 J and the impactor mass is 7.17 kg. The experimental impact behaviour has been assessed and discussed in terms of force, displacement, and energy responses.
After the impact, the specimens are subjected to non-destructive testing using micro‐CT. The internal damage mechanisms are
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors would like to thank the Ri-Ba Composites S.r.l. engineer for technical support and guidance throughout this research.
References (73)
- et al.
Predictive modeling of an energy-absorbing sandwich structural concept using the building block approach
Compos Appl Sci Manuf
(2010) - et al.
Process analysis for structural optimisation of thermoplastic composite component using the building block approach
Compos B Eng
(2017) - Furtado C, Catalanotti G, Arteiro A, Gray PJ, Wardle BL, Camanho PP. Simulation of failure in laminated polymer...
- Gemi L, Kayrıcı M, Uludağ M, Gemi DS, Şahin ÖS. Experimental and statistical analysis of low velocity impact response...
- et al.
Impact Response of Carbon/Kevlar Hybrid 3D Woven Composite Under High Velocity Impact: Experimental and Numerical Study
Appl Compos Mater
(2020) - et al.
Dynamic response of hybrid carbon fibre laminate beams under ballistic impact
Compos Struct
(2019) - Andrew JJ, Srinivasan SM, Arockiarajan A, Dhakal HN. Parameters influencing the impact response of fiber-reinforced...
- Choi HY, Chang fu K. A Model for Predicting Damage in Graphite/Epoxy Laminated Composites Resulting from Low-Velocity...
- et al.
Numerical modelling of impact damage
Composites
(1994) - et al.
Impact damage prediction in carbon composite structures
Int J Impact Eng
(1995)
Multiple delaminations and their severity in circular axisymmetric plates subjected to transverse loading
J Compos Mater
Shear driven delamination propagation in two dimensions
Compos Appl Sci Manuf
Mass criterion for wave controlled impact response of composite plates
Compos Appl Sci Manuf
Analytical prediction of large mass impact damage in composite laminates
Compos Appl Sci Manuf
Impact dynamics and damage in composite structures
Compos Struct
Investigation of two finite element modelling approaches for ballistic impact response of composite laminates
Int J Crashworthiness
Modelling damage evolution in composite laminates subjected to low velocity impact
Compos Struct
Finite element modelling of damage induced by low-velocity impact on composite laminates
Compos Struct
Modeling damage induced plasticity for low velocity impact simulation of three dimensional fiber reinforced composite
Compos Struct
Assessment of failure criteria and damage evolution methods for composite laminates under low-velocity impact
Compos Struct
Failure Criteria for unidirectional fiber composites
J Appl Mech
A General Theory of Strength for Anisotropic Materials
J Compos Mater
Failure analysis of FRP laminates by means of physically based phenomenological models
Compos Sci Technol
A continuous damage mechanics model for ductile fracture
J Eng Mater Technol
Review and benchmark study on the analysis of low-velocity impact on composite laminates
Eng Fail Anal
A numerical study on the impact resistance of composite shells using an energy based failure model
Compos Struct
Finite element analysis of a modified progressive damage model for composite laminates under low-velocity impact
Compos Struct
Impact Damage Characteristics of Carbon Fibre Metal Laminates: Experiments and Simulation
Appl Compos Mater
Simulating drop-weight impact and compression after impact tests on composite laminates using conventional shell finite elements
Int J Solids Struct
Simulation of low velocity impact on composite laminates with progressive failure analysis
Compos Struct
Numerical study for the structural analysis of composite laminates subjected to low velocity impact
Compos B Eng
Low-velocity impact predictions of composite laminates using a continuum shell based modeling approach Part b: BVID impact and compression after impact
Int J Solids Struct
Interlaminar interface modelling for the prediction of delamination
Compos Struct
Cited by (4)
Fatigue life simulation of the impacted carbon/epoxy composite laminates
2024, Composites Part A: Applied Science and ManufacturingFatigue and failure analysis of aluminium and composite automotive wheel rims: Experimental and numerical investigation
2023, Engineering Failure AnalysisLow-Speed Oblique Impact Response of Adhesively Bonded Dissimilar Single-Lap Joints
2022, Journal of Aerospace EngineeringOn the Numerical Modelling of Conductive CNT-Polymers: The Electro-mechanical Response
2022, Mechanisms and Machine Science