Curing‐Induced Warpage of 3D ‐Printed Continuous Carbon‐Fiber‐Reinforced Thermosetting Composites: Mechanisms, Machine‐Learning‐Assisted Analysis, and Parameter Optimization

ABSTRACT

3D‐printed continuous carbon‐fiber‐reinforced thermosetting composites (CCFRTCs) are widely researched for their excellent properties and design flexibility. However, curing‐induced warpage remains a major challenge, which severely restricts their engineering applications. Unlike traditional thermosetting composites, 3D‐printed CCFRTCs exhibit warpage that is highly sensitive to printing parameters. To explore the curing‐induced warpage of 3D‐printed CCFRTCs, a fully coupled thermo‐chemo‐mechanical model was developed. The numerically predicted warpage showed satisfactory consistency with experimental measurements, with deviations below 10% at most measurement points. Subsequently, key printing parameters, including printing temperature, printing speed, layer density, and printing spacing, were selected for systematic investigation. The results revealed that printing parameters affected the residual stress distribution and warpage primarily by altering the thickness of the interlayer neat resin region and overall structural stiffness. Furthermore, a machine learning (ML) model combined with SHapley Additive exPlanations (SHAP) interpretability analysis was employed to explore the multi‐parameter coupling effects on warpage. The results demonstrated that printing spacing showed the most prominent influence on warpage, followed by printing speed and layer density, whereas printing temperature exhibited the weakest effect. Finally, the particle swarm optimization (PSO) algorithm was adopted to optimize the printing parameters to minimize warpage, and the application validation was carried out through the fabrication of a 3‐m‐long wing spar with low warpage. These findings clarified the relationship between printing parameters and warpage and provided theoretical guidance for the engineering application of 3D‐printed CCFRTC components.