GEOMETRY PREDICTION SIMULATION FOR METAL ADDITIVE MANUFACTURING: A CALIBRATION-FREE APPROACH
Paper i proceeding, 2024
Metal additive manufacturing has significantly impacted the manufacturing industry, yet ensuring geometric quality remains a challenge during the design phase, with geometric distortion proving difficult to alleviate through post-treatment methods.In numerous manufacturing processes, simulations have proven effective; however, accurately simulating geometric outcomes in metal additive manufacturing necessitates the mathematical representation of various physical scales, spanning from the microscopic intricacies of powder and laser melting to the macroscopic deformation of components.Notably, existing simulation methods often fall short in accounting for the fluctuations in thermal history resulting from the cyclical heating and cooling inherent in layer-by-layer manufacturing processes.Also, as a design tool, there is a need for simple yet accurate simulations to aid the designer.This paper delves into the incorporation of thermal history variations and their influence on the accuracy of geometry predictions.This research introduces a new approach for using the inherent strain method.By employing micro-scale simulations, we conduct meso-scale simulations to estimate the plastic strain changes resulting from repeated heating-cooling cycles.The calculated plastic strain values are then used to determine inherent strains, which are applied layer by layer within block structures.This method considers the shrinkage experienced in previous layers, enabling accurate deformation simulations at the part scale.Validation of this approach involved a comparative analysis with five strain bridges, each featuring varying bridge thicknesses, fabricated through the Selective Laser Melting process.It is noted that the printed parts undergo a rotation after the support structure is cut off.When compensating for this rotation, the prediction of geometric distortion exhibited exceptional accuracy after cut-off.Consequently, the result suggests this method facilitates precise geometry simulation for components produced using the Selective Laser Melting process without the need for extensive inherent strain parameter tuning.The authors believe that this approach holds great promise as an aid for designers in the additive manufacturing industry.
Inherent strain
Selective laser melting
Additive manufacturing
Geometry assurance