Deep Learning-Based Cooperative Modelling of Thermal Flow and Stress Fields in Laser Powder Bed Fusion

In the realm of global manufacturing, the proliferation of Laser Powder Bed Fusion (LPBF) technologies has necessitated an in-depth understanding of the dynamics between thermal flow and stress fields during its operative procedures. Interactions between these fields have been observed to induce thermal distortions in components, potentially jeopardizing the structural integrity and operational efficiency of the end products. Insights have been garnered through conventional finite element methods and empirical models, yet these methodologies encounter evident constraints when deciphering highly nonlinear, multi-scale systems. This research delves into the employment of deep learning techniques for the cooperative modelling of the aforementioned fields, suggesting an innovative approach to thermal distortion predictions. The outcomes derived from this inquiry are foreseen to unveil novel optimization strategies for laser melting manufacturing methodologies, propelling the evolution of this specialized field.