A sensitivity analysis-based parameter optimization framework for 3D printing of continuous carbon fiber/epoxy composites

Three-dimensional printing of continuous carbon fiber/epoxy composites (CCF/EPCs) is an emerging additive manufacturing technology for fiber-reinforced polymer composites and has wide application prospects. However, the 3D printing parameters and their relationship with the mechanical properties of the final printed samples have not been fully investigated in a computational and quantifiable way. This paper presents a sensitivity analysis (SA)-based parameter optimization framework for the 3D printing of CCF/EPCs. A surrogate model for a process parameter-mechanical property relationship was established by support vector regression (SVR) analysis of the experimental data on flexural strength and flexural modulus under different process parameters. An SA was then performed on the SVR surrogate model to calculate the importance of each individual 3D printing parameter on the mechanical properties of the printed samples. Based on the SA results, the optimal 3D printing parameters and the corresponding flexural strength and flexural modulus of the printed samples were predicted and verified by experiments. The results showed that the proposed framework can serve as a high-accuracy tool to optimize the 3D printing parameters for the additive manufacturing of CCF/EPCs.