Abstract
Advanced composites have been increasingly used as primary load-bearing structures or structural members. While they offer many excellent properties, significant concerns regarding their long term durability and safety remain. One particular unsolved challenge is to quantify the uncertainty associated with the service life. Typical composites exhibit complex, multiple damage events including multiple crack initiation events within and between plies/tows from pristine materials, through their coupled growth, involving multiple bifurcation, coalescence, and ply-jumping, to eventual failure. Numerical and/or analytical methods capable of predicting the progressive damage evolution in composites under general thermal-mechanical loadings and quantifying their effects on structural integrity are in high demand.
High-fidelity thermal-mechanical analyses to composite structures with explicit consideration of arbitrary crack-like damage events are challenging because the heterogeneous nature of composites makes it impossible to know the cracking locations a priori, yet efficient accounting of the nucleation, coalescence, and bifurcation of multiple damage processes is critical to simulation fidelity. There are several methods that can deal with arbitrary cracking in solids such as the extended FEM (X-FEM) and the phantom-node-method (PNM). These methods typically employ crack tip enrichment functions through portioning-of-unity, which are inherently nonlocal and are difficult in handling merging or bifurcating cracks. The need for extra DoFs or nodes for new or propagating cracks also makes these methods inflexible in treating crack interaction.
In this study, we present a multiscale simulation platform based on our newly developed augmented finite element method (A-FEM). We shall demonstrate that the A-FEM is able to achieve high fidelity simulations of multiple, arbitrary damage evolution processes in composite structures with an improved numerical efficiency of several orders of magnitude. The capability of the new A-FEM will be demonstrated through several case studies on realistic composite systems at various important scales. Important information regarding how the microscopic damage processes and their evolution can impact the macroscopic structural performance/integrity will be discussed.
Abstract
The worlds of structural simulation and process simulation are rapidly converging as it becomes increasingly clear that failure predictions can be significantly enhanced if process-induced conditions and defects are captured in the structural simulation. Processing outcomes that affect structural response range from thermal outcomes (e.g. degree of cure or crystallinity, physical morphology), to quality outcomes (e.g. ply thickness, volume fraction, wrinkling and waviness, porosity), or mechanical outcomes (e.g. dimensional change and residual stress).
The state of the art in modelling these outcomes as a function of processing conditions is quite varied in terms of rigour, completeness, breadth, integration, and application. Broadly speaking, there are three ranges of matrix response where different simulations strategies can be applied: (a) high temperature, low cure/crystallinity where the system is essentially fluid like and viscous; (b) mid-temperature, gelled or crystallized where the system is viscoelastic; and (c) lower temperature, below the glass transition temperature, where the system is essentially elastic. The behaviour of the system must also be considered at all scales, which in processing includes not only the classical micro-scale, meso-scale, layer/lamina scale, and structural scale, but also the tooling, equipment and other associated manufacturing initial and boundary conditions. Here we present some of our most recent work in creating a coherent and consistent framework for process modelling, and provide experimental and simulation results to highlight both successes and needs. In particular, we focus on residual stress development under different thermal histories, and show how a multi-scale modelling approach can be exercised to evaluate and predict different aspects of this problem.
Abstract
Computational micromechanics is emerging as an accurate tool to study the mechanical
behaviour of composites at the micro scale due to the sophistication of the modeling tools and to the ever-increasing power of digital computers [1]. Within this framework, the macroscopic properties of a composite or heterogeneous material can be obtained by means of the numerical simulation of the deformation and failure of a statistically representative volume element of the microstructure. As compared with classic homogenization techniques, computational micromechanics presents two important advantages. Firstly, the influence of the geometry and spatial distribution of the phases (i.e. size, shape, clustering, connectivity, etc.) can be accurately taken into account. Secondly, the details of the stress and strain microfields throughout the microstructure are resolved, leading to precise predictions of the onset and propagation of damage.
Recent advances in this area include the analysis of the effect of different fiber (carbon and glass [2]), interfaces [3], spatial distribution and ply thickness or the influence of damage on the final mechanical performance of fiber-reinforced composites [4,5], as well as the physical determination by detailed experimental techniques based on instrumented nanoindentation [6,7] of the parameters governing the mechanical behavior of the composite at this length scale.
[1] Llorca, J; Gonzalez, C; Molina-Aldareguía, J M; Segurado, J; Seltzer, R; Sket, F; Rodriguez, M; Sadaba, S; Munoz, R; Canal, L P, Multiscale modeling of composite materials: a roadmap towards virtual testing, Advanced materials, 23, 5130-47, 2011, DOI: 10.1002/adma.201101683.
[2] E. Totry, J. M. Molina-Aldareguia, C. González, J. Llorca. Effect of fiber, matrix and interface properties on the inplane shear deformation of carbon-fiber reinforced composites. Composites Science and Technology, 70, 970-980, 2010.
[3] C. González and J. Llorca. Mechanical behaviour of unidirectional fiber-reinforced polymers under transverse compression: Microscopic mechanisms and modelling. Composites Science and Technology, 67, 2795-2806, 2007.
[4] LP. Canal, C. González, J. Segurado, J. LLorca, Intraply fracture of fiber-reinforced composites: Microscopic mechanisms and modeling, Composites Science and Technology, 72, 1223-1232,
DOI:10.1016/j.compscitech.2012.04.008, 2012.
[5] C. González and J. Llorca. Multiscale modelling of fracture in fiber-reinforced composites. Acta Materialia, 54, 4171-4181, 2006.
[6] M. Rodriguez, JM. Molina Aldareguia, C. González, J. LLorca, A methodology to measure the interface shear strength by means of the fiber push-in test, Composite Science and Technology, 72, 1924-1932,
DOI:10.1016/j.compscitech.2012.08.011, 2012.
[7] M. Rodriguez, JM. Molina-Aldareguia, C. González, J. LLorca, Determination of the mechanical properties of amorphous materials through instrumented nanoindentation, Acta Materialia, 60, 3953-3964,
DOI: 10.1016/j.actamat.2012.03.027, 2012.