Flattened yarn path point modeling method for complex 3D woven composites

Abstract To address challenges in accurately characterizing yarn paths and extracting thickness distributions for complex 3D woven composites in digital manufacturing, this paper proposes a parametric extraction method for yarn path points in the flattened state. First, the initial STL model is segmented along the warp direction according to weft width. Cross-sectional boundary points are extracted and interpolated clockwise to precisely partition each cross-sectional area. Then, a classification criterion is established to identify cross-sectional types, and a differentiated layering method is applied to extract initial yarn path points along the thickness direction. A flattening reference line is then determined using an algorithm adapted to each cross-sectional type. Finally, each layered cross-section is flattened based on this reference line, enabling the reconstruction of a 3D woven yarn path point model. The method was validated using 3D woven composite models with various structures and process parameters. Results demonstrate that the proposed approach accurately reconstructs yarn path points of complex preforms in the flattened state and accommodates diverse processing conditions, offering critical support for the digital design and manufacturing of 3D woven composites.