Simpleware Case Study: Direct Numerical Simulation of 3D Woven Textile Composites Subjected to Tensile Loading

Textile composites are widely used in industries like aerospace due to their structural properties, particularly their reinforcement in thickness directions. 3D woven textile composites also reduce manufacturing cost and time compared to traditional materials but require design and testing to better understand progressive damage and failure analysis. Traditionally, there have been challenges in multiscale modelling to understand the effect of the underlying microstructure and imperfections on failure. An image-based modelling approach using Simpleware software, along with NCYL multiscale code solved these problems by generating accurate meshes at different length scales, from the micro to the meso and macro-scale.

Overview

Textile composites are widely used in industries like aerospace due to their structural properties, particularly their reinforcement in thickness directions. 3D woven textile composites also reduce manufacturing cost and time compared to traditional materials but require design and testing to better understand progressive damage and failure analysis. Traditionally, there have been challenges in multiscale modelling to understand the effect of the underlying microstructure and imperfections on failure. An image-based modelling approach using Simpleware software, along with NCYL multiscale code solved these problems by generating accurate meshes at different length scales, from the micro to the meso and macro-scale.

Highlights

  • Multi-scale modeling method (NCYL Code) enables progressive damage and failure analysis of 3D woven textile composites
  • Modeling in Simpleware software from micro-CT allows study of microstructure imperfections in-situ
  • Results save computational resources and provide insights into high-value materials

Reference

Patel, D.K., Waas, A.M., Yen, C.F., 2018. Direct numerical simulation of 3D woven textile composites subjected to tensile loading: An experimentally validated multiscale approach. Composites Part B, Vol 152 (2018), pp. 102-115.

Thanks to

Deepak Patel, Department of Aerospace Engineering, University of Michigan, USA:

“I was looking for some solutions for how to handle these kinds of textile composites with a lot of imperfections, and how to capture these details in the model. I evaluated a few other softwares, and Simpleware served the purpose of what we’re looking for: it’s very robust…it gives good quality meshes.”

Simpleware Software for Multi-scale Meshing

New methods using Simpleware software result in high-fidelity and computationally accurate models across different length scales. Models are used to understand the relationship between the micro and macro scale at a full-coupon and Representative Volume Element (RVE) scale. Different SkyScan Micro-CT systems (1173; 1076; 1076) were used to obtain images at different length scales, from a 2” x 1” sample, to a half coupon, and full coupon. Half-coupon data was used for high-resolution manual segmentation in Simpleware software to reduce the size of the exported FE model for computational efficiency. Using Simpleware software, it was straightforward to segment individual fiber and matrix regions from the micro-CT data, and model in-situ imperfections.

Micro-CT scans at different length scales for a woven textile composite coupon

Comparison between micro-CT setting parameters of different scales

Each fiber tow in the image data was segmented and reconstructed into a 3D volume in Simpleware ScanIP. The Simpleware models from micro-CT reproduce the in-situ microstructure imperfections in the composite material.

Segmentation in Simpleware ScanIP of Micro-CT data of a woven textile composite coupon

Fibers segmentation in Simpleware ScanIP of tensile coupon

Simulation Results

Representative volume element (RVE) scanned image data was also meshed explicitly in Simpleware FE to capture geometric imperfections. The FE meshes were exported in native Abaqus format, including all finite element and nodal information along with the fiber path orientation. The fiber-matrix scale RVE was used to predict the fiber tow effective properties and the effect of micro-cracking and micro-damage in the composite’s polymer matrix. Inputs were then used to predict global stiffness and stresses in the macro-scale FEA from mesoscale stiffness tensors.

FE simulation of a woven textile composite coupon: two-piece failure due to weft fiber tows breakage

Two-piece failure due to weft fiber tows breakage

FE simulation of a woven textile composite coupon: final tensile failure

Final tensile failure of coupon

Tensile failure simulation of a woven textile composite coupon

Final tensile failure of coupon (3D isometric view)

Conclusions

An analytical subscale model (NCYL) also produced a computationally efficient framework for progressive damage analysis in the 3D woven textile composites. Simulation results of progressive damage and failure analysis agreed well with experimental data to understand in-plane elastic modulus in the weft direction. The global-local modelling strategy benefited from factoring in the in-situ microstructure imperfections for image-based modelling, while the subscale micromechanics model was able to predict the effective nonlinear response of a homogenized fiber tow. This method offers researchers lower computational cost and is suitable for studying large-scale progressive damage and failure analysis in high-value composites.

Any Questions?

Do you have any questions about this case study or how to use Simpleware software for your own workflows?