LAYER-WISE IN-PROCESS MONITORING-AND-FEEDBACK SYSTEM BASED ON SURFACE CHARACTERISTICS EVALUATED BY MACHINE-LEARNING-GENERATED CRITERIA

Access full-text files

Date

2023

Journal Title

Journal ISSN

Volume Title

Publisher

University of Texas at Austin

Abstract

In the laser powder bed fusion (PBF-LB) process, a set of parameters that are considered optimal are selected. Still, a set of parameters cannot accommodate complex model geometries, model placement in the build chamber, and unforeseen circumstances, leading to internal defects. Therefore, a new in-situ monitoring and feedback system has been developed to suppress the occurrence of lack-of-fusion (LOF) defects in the PBF-LB process. This system measures surface properties after each laser irradiation to predict whether LOF defects occur. Then, if necessary, a feedback process is performed to re-melt the same surface. Evaluation thresholds are defined by a combination of aerial surface texture parameters created in advance by machine learning of surface properties and defect occurrence. For example, a square pillar of Inconel 718 alloy built with feedback had a higher relative density than one without feedback.

Description

LCSH Subject Headings

Citation