Enhancing mechanical performance of metal additively manufactured structures via peridynamic topology optimization and thermo-mechanical process simulation with experimental validation

Residual stresses induced during metal additive manufacturing (AM) process can significantly compromise the structural performance of engineering parts, presenting challenges in achieving optimal mechanical properties. This study presents an integrated framework for residual stress minimization in metal AM, combining thermomechanical simulations with peridynamics-based topology optimization (PD-TO), and experimentally validates its effectiveness through AM production, surface distortion measurements, and mechanical testing. The PD-TO approach, known for its ability to capture structural discontinuities, is employed to optimize geometries. Partscale thermo-mechanical simulations are performed to calculate process-induced residual stresses, which are crucial for predicting potential crack initiation sites. The proposed methodology is demonstrated on two case studies: a three-dimensional cantilever beam and an engine mounting bracket. The framework focuses on design modifications based on residual stress predictions, while keeping process parameters fixed during simulation and manufacturing. While the first case presents purely numerical results, the second integrates computational results with experimental validations. The numerical results reveal a substantial reduction in residual stress, and the experimental findings further support this improvement-demonstrating a 43 % weight reduction while preserving strain energy levels and minimizing distortions. These results underscore the potential of the PD-TO framework to enhance the design of metal AM parts by effectively addressing residual stress-induced defects.