Peridynamic Modeling of Process-Dependent Material Strength in Additive Manufacturing

Kurzfassung

Additive Manufacturing (AM) processes create complex, process-dependent material properties during printing that cannot be measured in advance and are governed by parameters such as tool path, temperature history, and cooling conditions. A Peridynamics (PD) correspondence framework is presented for simulating the entire AM process chain - printing, cooling, and mechanical loading - to predict the resulting material strength. Polymer crystallization of Polyether Ether Ketone (PEEK) captured via a dual-kinetic model is implemented as a subroutine compatible with the Abaqus HETVAL interface; the resulting crystallinity governs the local nodal stiffness. Thermo-mechanical coupling with convective heat transfer is handled within the PD correspondence formulation. A matrix-based quasi-static solver is coupled with an explicit Velocity-Verlet integrator for the fracture phase to eliminate the Courant-Friedrichs-Lewy (CFL) bottleneck of purely dynamic integration. The framework is validated on PEEK dogbone specimens (ASTM D638) for four environment temperatures, demonstrating qualitatively correct crack initiation and propagation and a speedup exceeding two orders of magnitude over a fully dynamic reference computation.