Integrated computational modeling of large format additive manufacturing: Developing a digital twin for material extrusion with carbon fiber-reinforced acrylonitrile butadiene styrene

Abstract

Material extrusion (MEX) continues to be a pivotal additive manufacturing (AM) process, involving the selective heating and layer-by-layer deposition of material. However, conventional finite-element models (finite-element analysis) face limitations in accurately simulating the MEX process, highlighting the need for experimental validation. This paper highlights an advanced material modeling technology that streamlines the development of composite parts using large format AM (LFAM). It specifically focuses on a thermoplastic acrylonitrile butadiene styrene (ABS) matrix composite material enriched with 20% short carbon fibers. The study employs integrated computational materials engineering, integrating (i) the manufacturing process, (ii) the material’s microstructure, (iii) homogenization techniques, and (iv) the performance of the final part. The development of a digital twin for pellet extrusion is proposed, emphasizing the importance of micro-structure characterization to account for warpage and residual stresses that lead to part distortion. The demonstrator manufactured for this study is a wind turbine mold of a blade section. Experimental tests revealed an elastic modulus of 5.5 GPa and a hardening modulus of 2.4 GPa for the composite. The numerical microscopic model showed a 16% error in the elastic modulus compared to experimental results. The study concludes that the homogenization techniques are effective in predicting the elastic properties but lack accuracy in the plastic region. The application of the model to the LFAM process of pellet extrusion is demonstrated, with simulation results showing a maximum deformation close to the center of the wind turbine blade mold.