AI-driven process monitoring enhances the qualification of additively manufactured stainless-steel parts. By analysing high-frequency welding current and voltage signals in both the time and frequency domains, the proposed Isolation Forest–based method (IMAGE)
Caption
AI-driven process monitoring enhances the qualification of additively manufactured stainless-steel parts. By analysing high-frequency welding current and voltage signals in both the time and frequency domains, the proposed Isolation Forest–based method detects anomalies with 85.3% accuracy—outperforming traditional Statistical Process Monitoring, enabling preventive quality assessment, reducing inspection time and production costs while improving reliability in Wire Arc Direct Energy Deposition processes.
Credit
Giulio Mattera/University of Naples “Federico II”, Zengxi Pan/University of Wollongong, Elena Manoli/University of Naples “Federico II”, Luigi Nele/University of Naples “Federico II”
Usage Restrictions
Credit must be given to the creator.
License
CC BY