Ingenieurwissenschaften und zugeordnete Tätigkeiten
Filtern
Dokumenttyp
- Beitrag zu einem Tagungsband (1)
- Vortrag (1)
Sprache
- Englisch (2)
Referierte Publikation
- nein (2)
Schlagworte
- In-situ monitoring (2) (entfernen)
Organisationseinheit der BAM
Eingeladener Vortrag
- nein (1)
In this study, we present an enhanced deep learning framework for the prediction of porosity based on thermographic in-situ monitoring data of laser powder bed fusion processes. The manufacturing of two cuboid specimens from Haynes 282 (Ni-based alloy) powder was monitored by a short-wave infrared camera. We use thermogram feature data and x-ray computed tomography data to train a convolutional neural network classifier. The classifier is used to perform a multi-class prediction of the spatially resolved porosity level in small sub-volumes of the specimen bulk.
Laser Powder Bed Fusion (L-PBF), as one of the most promising production process in the field of metal additive manufacturing, enables traditional constructive solutions to be rethought and the manufacturing of optimized components according to the "form follows function" principle. The most significant obstacle for a broad industrial application of the L-PBF process is the inadequate quality assurance during the manufacturing process so far, leading to high production costs. Although several mainly camera based commercial in-process monitoring systems are already available, a deep understanding of the interpretation of the monitored data and correlation with actual defects is still lacking. One reason for this is the reduction of the complex process signature to just one measurement value.
The focus of this contribution is the presentation of the multispectral optical tomography as alternative to single measurand in-situ monitoring systems. The potential of this approach is hereby shown on L-PBF printed samples with induced process instabilities. Beyond that, an in-house developed L-PBF printer for further testing of multi-sensor in-situ monitoring systems is presented.