Filtern
Erscheinungsjahr
- 2023 (3) (entfernen)
Dokumenttyp
- Vortrag (2)
- Beitrag zu einem Tagungsband (1)
Sprache
- Englisch (3)
Referierte Publikation
- nein (3)
Schlagworte
- In-situ monitoring (3) (entfernen)
Organisationseinheit der BAM
Eingeladener Vortrag
- nein (2)
In recent years, additive manufacturing technologies have gained in importance. Laser powder bed fusion can be used for complex functional components or the production of workpieces in small quantities. High safety requirements, e.g. in aerospace, demand comprehensive quality control. Therefore, non-destructive offline inspection methods such as computed tomography are used after production. Recently, online non-destructive testing methods such as optical tomography have been developed to improve profitability and practicality. In this presentation, the applicability of eddy current inspection using GMR sensors for online inspection of PBF-LB/M parts is demonstrated. Eddy current testing is performed for each layer during the production process at frequencies uo to 1.2 MHz. Despite the use of high-resolution arrays with 128 elements, the testing time is kept low by an adapted hardware. Thus, the measurement can be performed during the manufacturing process without significantly slowing down the production process. In addition to the approach, the results of an online eddy current test of a step-shaped test specimen made of Haynes282 are presented.
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.
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.