Wissenschaftliche Artikel der BAM
Filtern
Erscheinungsjahr
- 2023 (4) (entfernen)
Dokumenttyp
Sprache
- Englisch (4)
Schlagworte
- Additive manufacturing (3)
- Thermography (2)
- BPF-LB/M (1)
- Convolutional neural networks (CNN) (1)
- Defect detection (1)
- Flaw detection (1)
- Heat accumulation (1)
- In situ monitoring (1)
- In-situ (1)
- Laser Powder Bed Fusion (PBF-LB/M, L-PBF) (1)
Organisationseinheit der BAM
- 8.0 Abteilungsleitung und andere (4) (entfernen)
Laser powder bed fusion of metallic components (PBF-LB/M) is gaining acceptance in industry. However, the high costs and lengthy qualification processes required for printed components create the need for more effective in-situ monitoring and testing methods. This article proposes multispectral Optical Tomography (OT) as a new approach for monitoring the PBF-LB/M process. Compared to other methods, OT is a low-cost process monitoring method that uses long-time exposure imaging to observe the build process. However, it lacks time resolution compared to expensive thermographic sensor systems. Monochromatic OT (1C-OT) is already commercially available and observes the building process layer-wise using a single wavelength window in the NIR range. Multispectral OT (nC-OT) utilizes a similar setup but can measure multiple wavelength ranges per location simultaneously. By comparing the classical 1C-OT and nC-OT approaches, this article examines the advantages of nC-OT (two channel OT and RGB-OT) in reducing the false positive rate for process deviations and approximating maximum temperatures for a better comparison between different build processes and materials. This could ultimately reduce costs and time for part qualification. The main goal of this contribution is to assess the advantages of nC-OT compared to 1C-OT for in-situ process monitoring of PBF-LB/M.
The capability to produce complexly and individually shaped metallic parts is one of the main advantages of the laser powder bed fusion (PBF LB/M) process. Development of material and machine specific process parameters is commonly based on results acquired from small cubic test coupons of about 10 mm edge length. Such cubes are usually used to conduct an optimization of process parameters to produce dense material. The parameters are then taken as the basis for the manufacturing of real part geometries. However, complex geometries go along with complex thermal histories during the manufacturing process, which can significantly differ from thermal conditions prevalent during the production of simply shaped test coupons. This may lead to unexpected and unpredicted local inhomogeneities of the microstructure and defect distribution in the final part and it is a root cause of reservations against the use of additive manufacturing for the production of safety relevant parts. In this study, the influence of changing thermal conditions on the resulting melt pool depth of 316L stainless steel specimens is demonstrated. A variation of thermo-graphically measured intrinsic preheating temperatures was triggered by an alteration of inter layer times and a variation of cross section areas of specimens for three distinct sets of process parameters. Correlations between the preheating temperature, the melt pool depth, and occurring defects were analyzed. The limited expressiveness of the results of small density cubes is revealed throughout the systematic investigation. Finally, a clear recommendation to consider thermal conditions in future process parameter optimizations is given.
Great complexity characterizes Additive Manufacturing (AM) of metallic components via laser powder bed fusion (PBF-LB/M). Due to this, defects in the printed components (like cracks and pores) are still common. Monitoring methods are commercially used, but the relationship between process data and defect formation is not well understood yet. Furthermore, defects and deformations might develop with a temporal delay to the laser energy input. The component’s actual quality is consequently only determinable after the finished process.
To overcome this drawback, thermographic in-situ testing is introduced. The defocused process laser is utilized for nondestructive testing performed layer by layer throughout the build process. The results of the defect detection via infrared cameras are shown for a research PBF-LB/M machine.
This creates the basis for a shift from in-situ monitoring towards in-situ testing during the AM process. Defects are detected immediately inside the process chamber, and the actual component quality is determined.
The prediction of porosity is a crucial task for metal based additive manufacturing techniques such as laser powder bed fusion. Short wave infrared thermography as an in-situ monitoring tool enables the measurement of the surface radiosity during the laser exposure. Based on the thermogram data, the thermal history of the component can be reconstructed which is closely related to the resulting mechanical properties and to the formation of porosity in the part. In this study, we present a novel framework for the local prediction of porosity based on extracted features from thermogram data. The framework consists of a data pre-processing workflow and a supervised deep learning classifier architecture. The data pre-processing
workflow generates samples from thermogram feature data by including feature information from multiple subsequent layers.
Thereby, the prediction of the occurrence of complex process phenomena such as keyhole pores is enabled. A custom convolutional neural network model is used for classification. Themodel is trained and tested on a dataset from thermographic in-situ monitoring of the manufacturing of an AISI 316L stainless steel test component. The impact of the pre-processing parameters and the local void distribution on the classification performance is studied in detail. The presented model achieves an accuracy of 0.96 and an f1-Score of 0.86 for predicting keyhole porosity in small sub-volumes with a dimension of (700 × 700 × 50) μm3. Furthermore, we show that pre-processing parameters such as the porosity threshold for sample
labeling and the number of included subsequent layers are influential for the model performance. Moreover, the model prediction is shown to be sensitive to local porosity changes although it is trained on binary labeled data that disregards the actual sample porosity.