08 - Manufacturing and Supply Chain Management

A NOVEL DUAL-PHASE BAYESIAN ESTIMATION AND MULTI-TASK LEARNING METHOD FOR PROCESS OPTIMIZATION FOR REDUCING LACK-OF-FUSION DEFECTS DURING LASER POWDER BED FUSION

Y.S. Chen¹, D.D. Gu¹; ¹Nanjing University of Aeronautics and Astronautics, China

We presented a dual-phase Bayesian estimation and physics-informed multi-task learning framework to suppress lack-of-fusion in laser powder bed fusion aeronautical parts. Multi-track melt-pool geometry and dimensionless criteria build lack-of-fusion sensitivity maps, revealing robust power–speed windows with minimal experiments for reliable aeronautical components.

Sorry, there is no file available for this paper.