Additive construction techniques, such as 3D Concrete Printing (3DCP), offer significant advantages, as reduced material waste, optimized construction times, and the possibility to create complex shapes that would be difficult to achieve through conventional methods. Despite the advances in 3DCP, some challenges persist due to the still incipient development to predict the printing results. The fresh properties of the material play a crucial role. Therefore, in addition to experimental studies, numerical modeling emerges as a valuable tool to investigate the dynamic aspects of the extrusion process in 3DCP. However, the numerical modeling of 3DCP is challenging due to the complex rheological behavior of the self-supporting concrete, which typically exhibits high values of apparent viscosity at low deformation rates. To address these challenges, we adopted the moving particle semi-implicit (MPS) method, which is a particle-based simulation technique suitable for modeling free surface flows and materials undergoing large deformations. As a Lagrangian approach, MPS discretizes the computational domain into particles that move according to the governing equations of fluid motion. The conventional formulation of the MPS presents numerical restrictions when dealing with highly viscous fluids. To assure numerical stability, the semi-implicit approach requires very low time steps, resulting in high processing costs. In this way, this study adopts an implicit algorithm that allows for significantly larger time steps, resulting in reduced processing time compared to the conventional semi-implicit formulation. To validate the implicit algorithm, simulations of non-Newtonian flow between parallel plates were validated using analytical solutions, demonstrating better accuracy at lower Bingham numbers. The 3DCP simulations were conducted to describe the flow of fresh mortar using the Bingham-Papanastasiou constitutive model. The geometry of the 3D model includes the extrusion nozzle and a planar surface where concrete is printed. The numerical investigation involved simulations with different nozzle heights (Hn), printing speeds (V), and extrusion volumetric fluxes (U). The cross-sectional shapes of the extruded layers were compared with experimental data, showing qualitative agreement. The results demonstrate the potential of the implicit MPS method for modeling 3DCP processes.
Read full abstract