Contribution
Lecture shiftet to Thu, 10h00 in Num. Ac. 2: Insights Into Automated Quality Inspection in Lightweight Additive Manufacturing Using Machine-Learning With Measured and Simulated Acoustic Emissions
* Presenting author
Abstract:
Additive manufacturing through laser-based powder bed fusion offers unprecedented flexibility in designing complex and lightweight components. However, the process is susceptible to defects forming during the melting and solidification process, which impacts the final quality of the product. To address these challenges, the research project ML-S-LeAF introduces automated quality assurance by integrating machine learning models trained onacoustic emissions data. Structure-borne as well as airborne sound, which have both been obtained through in situ sensor measurements and acoustics simulations, are analyzed with respect to their efficacy in reliably detecting anomalies linked to defects in the printing process. Our final results confirm that the proposed machine learning approach can detect defects of printed lines with 98% accuracy. The integration of data-driven monitoringsets a future foundation for achieving consistent and reliable quality in additive manufacturing for lightweight construction.