Litcius/Paper detail

Rapidly predicting the effect of tool geometry on the wrinkling of biaxial NCFs during composites manufacturing using a deep learning surrogate model

J.V. Viisainen, F. Richard Yu, A. Codolini, Shuai Chen, L.T. Harper, M.P.F. Sutcliffe

2023Composites Part B Engineering34 citationsDOIOpen Access PDF

Abstract

A deep learning surrogate model is developed to rapidly predict the wrinkling patterns of a biaxial non-crimp fabric (NCF) layup for any given tool geometry during forming. The underlying dataset of finite element simulations is used to investigate the effect of tool geometry on wrinkling severity. The trained surrogate model is able to make reliable predictions of wrinkling patterns at a very low computational cost, suitable for tool design optimisation. Results indicate that certain geometrical features have a greater impact on wrinkling than others. In particular, forming NCFs over geometries with greater draft angles tends to result in smaller wrinkles.

Topics & Concepts

CrimpFinite element methodStructural engineeringGeometrySurrogate modelMaterials scienceBucklingComputer scienceComposite materialMathematicsEngineeringMachine learningStructural Analysis and OptimizationTextile materials and evaluationsMechanical Behavior of Composites
Rapidly predicting the effect of tool geometry on the wrinkling of biaxial NCFs during composites manufacturing using a deep learning surrogate model | Litcius