Ingenieurwissenschaften und zugeordnete Tätigkeiten
Filtern
Dokumenttyp
- Vortrag (3)
Referierte Publikation
- nein (3)
Schlagworte
- Defect detection (3) (entfernen)
Organisationseinheit der BAM
Eingeladener Vortrag
- nein (3) (entfernen)
Laser powder bed fusion is one of the most promising additive manufacturing techniques for printing complex-shaped metal components. However, the formation of subsurface porosity poses a significant risk to the service lifetime of the printed parts. In-situ monitoring offers the possibility to detect porosity already during manufacturing. Thereby, process feedback control or a manual process interruption to cut financial losses is enabled.
Short-wave infrared thermography can monitor the thermal history of manufactured parts which is closely connected to the probability of porosity formation. Artificial intelligence methods are increasingly used for porosity prediction from the obtained large amounts of complex monitoring data. In this study, we aim to identify the potential and the challenges of deep-learning-assisted porosity prediction based on thermographic in-situ monitoring.
Therefore, the porosity prediction task is studied in detail using an exemplary dataset from the manufacturing of two Haynes282 cuboid components. Our trained 1D convolutional neural network model shows high performance (R2 score of 0.90) for the prediction of local porosity in discrete sub-volumes with dimensions of (700 x 700 x 40) μm³.
It could be demonstrated that the regressor correctly predicts layer-wise porosity changes but presumably has limited capability to predict differences in local porosity. Furthermore, there is a need to study the significance of the used thermogram feature inputs to streamline the model and to adjust the monitoring hardware. Moreover, we identified multiple sources of data uncertainty resulting from the in-situ monitoring setup, the registration with the ground truth X-ray-computed tomography data and the used pre-processing workflow that might influence the model’s performance detrimentally.
The characterisation of AM structures is an important aspect of the AM process, required in order to:
1. optimise the AM printing process
2. assess the quality of produced parts
A wide range of characterisation techniques are available, and the selection can be complex, based on multiple factors.
One output from the MetAMMi project is a good practice guide on the correct choice of characterisation technique.
The detection of delamination, rot, and cracks in wooden construction elements is a key task for nondestructive testing (NDT). Air-coupled ultrasound (ACU) is used to detect defects and damage without altering the structure permanently. Using through transmission it is possible to detect even small holes and missing adhesive. After interpretation of the inspection data, an assessment of the mechanical properties based on an appraisal of internal defects in the material is feasible. Novel ferroelectret transducers with a high signal-to-noise ratio (SNR) enable a high-precision structure recognition. The transducers made of cellular polypropylene (PP) are quite suitable for ACU testing due to their extremely low Young’s modulus and low density which result in a favorable acoustic impedance for the transmission of ultrasonic waves between the transducer and air. Thus, structures with great dimensions, thickness up to 300 mm and material densities under 500 kg/m³ can be inspected. Promising results were obtained under laboratory conditions with frequencies from 90 kHz to 200 kHz. The advantage of ACU transducers is that they do not require contact to the sample, are accurate, and cost effective. Ultrasonic quality assurance for wood is an important attempt to increase the acceptance of wooden structures and towards sustainability in civil engineering in general.