Search

The application of 3D weaves has advantages over conventional uni-directional or 2D woven lay-ups. There is potential to produce near net-shaped preforms and to increase damage resistance due to the presence of through thickness reinforcement. Conventional 3D weaves typically consist of orthogonal yarns interwoven with through thickness binder yarns. This paper describes a feasibility study to find optimum architectures for 3D woven fabrics where some of the normal manufacturing constraints are relaxed. This will provide the basis for development of novel manufacturing methods based on optimum textile architectures. A framework has been developed for the automatic generation and analysis of 3D textile geometries, utilising the open-source pre-processor TexGen. A genetic algorithm is used to select an optimum geometry by evaluating results from finite element simulations using the commercial solver Abaqus. This paper highlights the flexibility of TexGen software to create complex 3D models by means of its Python scripting application programming interface (API). A standard layer-to-layer geometry is used as a starting point to which off-axis yarn rotations, in-plane shift of entire layers and adjustments to binder yarns can be applied. Geometric variables are selected to represent the textile architecture enabling the automation of unit cell creation and finite element analysis. A Genetic Algorithm is used to determine the optimum through thickness binder path, the number and the width of the binders, and yarn angles using a weighted objective function of the material elastic properties. The case studies show that the algorithm is efficient to converge to the optimum fibre architecture.

An experimental and numerical investigation was undertaken to characterise the collapse of 3D orthogonal woven carbon fibre composites during the load cases of in-plane tension, inplane compression and out-of-plane bending. Two different fibre architectures, varying only by the density of through-thickness reinforcement, were investigated. Cantilever beam tests were carried out to isolate two distinct collapse mechanisms, i.e. bending governed and shear governed deformation. A qualitative comparison was made with a similar UD-laminate material. 3D woven composites exhibited significantly reduced delamination. An investigation into the efficacy of an embedded element modelling strategy for in-plane tension, in-plane compression and out-of-plane bending load cases was undertaken. The predictions were generally in good agreement with the experimental measurements for the inplane and out-of-plane loading.

There are vast possibilities in fibre architecture design of 3D woven reinforcement. This paper considers the application of Genetic Algorithm (GA) in 3D woven composites optimisation. A set of real and integral variables, representing 3D fibre architecture, are formulated into a mixed integer Genetic Algorithm. The objective function is evaluated through automation of the unit cell based finite element analysis, by using the open source pre-processor TexGen and the commercial solver ABAQUS. The mixed integer Genetic Algorithm is adapted to a micro-population, aiming to improve computational efficiency. The study uses statistical tests to quantify the performance of the Genetic Algorithm schemes and the choice of parameters. The proposed approach was applied to the optimisation of 3D woven composites for maximum buckling resistance for the case of a landing gear brace. This study demonstrated that the optimisation converged to the optimum design within 20 iterations, considering 300 out of 7000 permissible solutions. In terms of buckling performance, the optimum design performed twice as well as cross-ply laminated composites and at least 50% better than known orthogonal 3D woven composites.

This paper presents an experimental and numerical investigation into the dynamic response of 3D orthogonal woven carbon composites undergoing soft impact. Composite beams of two different fibre architectures, varying only by the density of through-thickness reinforcement, were centrally impacted by metallic foam projectiles. Using high speed photography, the centre-point back-face deflection was measured as a function of projectile impulse. Qualitative comparisons are made with a similar uni-directional laminate material. No visible delamination occurred in orthogonal 3D woven samples, and beam failure was caused by tensile fibre fracture at the gripped ends. This contrasts with uni-direction carbon fibre laminates, which exhibit a combination of wide-spread delamination and tensile fracture. Post-impact clamped-clamped beam bending tests were undertaken across the range of impact velocities tested in order to investigate any internal damage within the material. Increasing impact velocity caused a reduction of beam stiffness: this phenomenon was more pronounced in composites with a higher density of through-thickness reinforcement. A three-dimensional finite element modelling strategy is presented and validated, showing excellent agreement with the experiment in terms of back-face deflection and damage mechanisms. The numerical analyses confirm negligible influence from though-thickness reinforcement in regards to back-face deflection, but significant reductions in delamination damage propagation. Finite element modelling was used to demonstrate the significant structural enhancements provided by the through-the-thickness weave. The contributions to the field made by this research include the characterisation of 3D woven composite materials under high-speed soft impact, and the demonstration of how established finite element modelling methodologies can be applied to the simulation of orthogonal woven textile composite materials undergoing soft impact loading.

Manipulation of the through-thickness yarn path enables 3D woven reinforcement to separate locally in the form of a bifurcation, creating net-shaped preforms for T- and I-beams. Preforming introduces fibre architecture deformation at the 3D woven bifurcation area. We report a geometric modelling approach to represent the realistic fibre architecture, as a preprocessing tool for finite element analyses. The study started with x-ray micro-computed tomography (µCT) of two 3D woven T-beams varying only by their yarn path at the T-junction area. Supported by the µCT image analysis, a set of mathematical formula were proposed to describe the identified features in the 3D woven T-beams. We then moved on to implement the automated modelling procedure in the open-source software TexGen. Using the weave pattern as input data, TexGen first simulates as-woven flat T-piece. Next, TexGen applies geometric transformation and refinements to simulate the preforming process of T-beams. The paper highlights an efficient approach to model the complex woven bifurcation structure at mesoscale.

The study focuses on unit cell FE modelling to predict coefficients of thermal expansion (CTEs) for sheared fabric laminates. Shear, as a dominant deformation mode in textile composites forming, introduces high degrees of anisotropy in both elasticity and thermal expansion. The unit cell predictions are based on realistic fibre architecture and measured material properties of constituent fibre and resin. Under the multi-scale framework, the unit cell predictions are part of the essential input data for locally varied material definitions. These definitions are used to model structural components to predict shape distortion. The FE model gives predictions close to the experimental data, when the boundary conditions are correlated to the coupon size. Nesting is an influential factor for CTEs. For true material representation, in-plane periodicity and nesting have been considered.

Numerical methods have become increasingly effective tools for analysis and design of composite materials. This study investigates how the inclusion of geometrical variations in modelling 3D woven fabrics affects the accuracy of numerical predictions. Based on micro-Computed Tomography data of 3D orthogonal woven composites, unit cell models were generated in TexGen at different levels of geometrical detail. Two types of analysis were implemented: (a) computational fluid dynamics (CFD) simulates resin flow during fabric impregnation in composites processing to predict permeability; (b) implicit static finite element analysis predicts in-plane tensile strength of the composites. By comparison with experimental data, the numerical predictions indicate that local geometrical variations, particularly in yarn cross-section, surface crimp and binder yarn path, have significant influence on both permeability and material strength. It is important to model the precise geometry in certain locations while the overall geometry can be simplified in order to maintain the practicality of model generation.

The advance of 3D fabric technology allows tailored material structure in different directions for optimised performance. 3D fabrics open up increasing applications in automotive, medical, energy and many other areas. This paper explores highly automated techniques to simulate 3D fabric geometry and mechanical behaviour. The basis of the work starts from TexGen,an open source software package developed at the University of Nottingham. A complex variety of 3D fabrics can be defined as subclass functions from base functions for individual yarns, crosssections and yarn paths. The 3D fabric geometry can be generated automatically in TexGen using a script given a number of parameters. From this geometrical model, an automated procedure is followed to create an input file for finite element analysis. Yarns are meshed with hexahedral and wedge elements. The input file contains the mesh, element orientations, material definitions, contact surfaces and definitions, periodic boundary conditions and loading steps. Once the FEA simulation is completed, the nodal displacement data are parsed into TexGen to build the deformed 3D fabric geometry. The deformed geometry model can then be used for further analyses, eg. to predict permeability or composite mechanical properties. An example for a complex three layer woven fabric is demonstrated.