Ingenieurwissenschaften und zugeordnete Tätigkeiten
Filtern
Erscheinungsjahr
- 2022 (8) (entfernen)
Dokumenttyp
- Vortrag (6)
- Zeitschriftenartikel (2)
Sprache
- Englisch (8) (entfernen)
Schlagworte
- Thermography (8) (entfernen)
Organisationseinheit der BAM
- 8 Zerstörungsfreie Prüfung (8) (entfernen)
Eingeladener Vortrag
- nein (6)
The formation of irregularities such as keyhole porosity pose a major challenge to the manufacturing of metal parts by laser powder bed fusion (PBF-LB/M). In-situ thermography as a process monitoring technique shows promising potential in this manner since it is able to extract the thermal history of the part which is closely related to the formation of irregularities. In this study, we investigate the utilization of machine learning algorithms to detect keyhole porosity on the base of thermographic features. Here, as a referential technique, x-ray micro computed tomography is utilized to determine the part's porosity. An enhanced preprocessing workflow inspired by the physics of the keyhole irregularity formation is presented in combination with a customized model architecture. Furthermore, experiments were performed to clarify the role of important parameters of the preprocessing workflow for the task of defect detection . Based on the results, future demands on irregularity prediction in PBF-LB/M are derived.
The appearance of irregularities such as keyhole porosity is a major challenge for the production of metal parts by laser powder bed fusion (PBF-LB/M). The utilization of thermographic in-situ monitoring is a promising approach to extract the thermal history which is closely related to the formation of irregularities. In this study, we investigate the utilization of convolutional neural networks to predict keyhole porosity based on thermographic features. Here, the porosity information calculated from an x-ray micro computed tomography scan is used as reference. Feature engineering is performed to enable the model to learn the complex physical characteristics of the porosity formation. The model is examined with regard to the choice of hyperparameters, the significance of thermal features and characteristics of the data acquisition. Based on the results, future demands on irregularity prediction in PBF-LB/M are derived.
The formation of defects such as keyhole pores is a major challenge for the production of metal parts by Laser Powder Bed Fusion (LPBF). The LPBF process is characterized by a large number of influencing factors which can be hard to quantify. Machine Learning (ML) is a prominent tool to predict the outcome of complex processes on the basis of different sensor data. In this study, a ML model for defect prediction is created using thermographic image features as input data. As a reference, the porosity information calculated from an x-ray Micro Computed Tomography (µCT) scan of the produced specimen is used. Physical knowledge about the keyhole pore formation is incorporated into the model to increase the prediction accuracy. From the prediction result, the quality of the input data is evaluated and future demands on in-situ monitoring of LPBF processes are formulated.
Additive manufacturing (AM) techniques have risen to prominence in many industrial sectors. This rapid success of AM is due to the freeform design, which offers enormous possibilities to the engineer, and to the reduction of waste material, which has both environmental and economic advantages. Even safety-critical parts are now being produced using AM. This enthusiastic penetration of AM in our daily life is not yet paralleled by a thorough characterization and understanding of the microstructure of materials and of the internal stresses of parts. The same holds for the understanding of the formation of defects during manufacturing. While simulation efforts are sprouting and some experimental techniques for on-line monitoring are available, still little is known about the propagation of defects throughout the life of a component (from powder to operando/service conditions). This Issue was aimed at collecting contributions about the advanced characterization of AM materials and components (especially at large-scale experimental facilities such as Synchrotron and Neutron sources), as well as efforts to liaise on-line process monitoring to the final product, and even to the component during operation. The goal was to give an overview of advances in the understanding of the impacts of microstructure and defects on component performance and life at several length scales of both defects and parts.
Avoiding the formation of defects such as keyhole pores is a major challenge for the production of metal parts by Laser Powder Bed Fusion (LPBF). The use of in-situ monitoring by thermographic cameras is a promising approach to detect defects, however the data is hard to analyze by conventional algorithms. Therefore, we investigate the use of Machine Learning (ML) in this study, as it is a suitable tool to model complex processes with many influencing factors. A ML model for defect prediction is created based on features extracted from process thermograms. The porosity information calculated from an x-ray Micro Computed Tomography (µCT) scan is used as reference. Physical characteristics of the keyhole pore formation are incorporated into the model to increase the prediction accuracy. Based on the prediction result, the quality of the input data is inferred and future demands on in-situ monitoring of LPBF processes are derived.
The industrial use of additive manufacturing for the production of metallic parts with high geometrical complexity and lot sizes close to one is rapidly increasing as a result of mass individualisation and applied safety relevant constructions. However, due to the high complexity of the production process, it is not yet fully understood and controlled, especially for changing (lot size one) part geometries.
Due to the thermal nature of the Laser-powder bed fusion (L-PBF) process – where parts are built up layer-wise by melting metal powder via laser - the properties of the produced part are strongly governed by its thermal history. Thus, a promising route for process monitoring is the use of thermography. However, the reconstruction of temperature information from thermographic data relies on the knowledge of the surface emissivity at each position on the part. Since the emissivity is strongly changing during the process due to phase changes, great temperature gradients, possible oxidation, and other potential influencing factors, the extraction of real temperature data from thermographic images is challenging. While the temperature development in and around the melt pool, where melting and solidification occur is most important for the development of the part properties. Also, the emissivity changes are most severe in this area, rendering the temperature deduction most challenging.
A possible route to overcome the entanglement of temperature and emissivity in the thermal radiation is the use of hyperspectral imaging in combination with temperature emissivity separation (TES) algorithms. As a first step towards the combined temperature and emissivity determination in the L-PBF process, here, we use a hyperspectral line camera system operating in the short-wave infrared region (0.9 µm to 1.7 µm) to measure the spectral radiance emitted. In this setup, the melt pool of the L-PBF process migrates through the camera’s 1D field of view, so that the radiation intensities are recorded simultaneously for multiple different wavelength ranges in a spatially resolved manner. At sufficiently high acquisition frame rate, an effective melt pool image can be reconstructed. Using the grey body approximation (emissivity is independent of the wavelength), a first, simple TES is performed, and the resulting emissivity and temperature values are compared to literature values. Subsequent work will include reference measurements of the spectral emissivity in different states allowing its analytical parametrisation as well as the adaption and optimisation of the TES algorithms. An illustration of the proposed method is shown in Fig.1.
The investigated method will allow to gain a deeper understanding of the L-PBF process, e.g., by quantitative validation of simulation results. Additionally, the results will provide a data basis for the development of less complex and cheaper sensor technologies for L-PBF in-process monitoring (or for related process), e.g., by using machine learning.
Additive manufacturing (AM) technologies, generally called 3D printing, are widely used because their use provides a high added value in manufacturing complex-shaped components and objects. Defects may occur within the components at different time of manufacturing, and in this regard, non-destructive techniques (NDT) represent a key tool for the quality control of AM components in many industrial fields, such as aerospace, oil and gas, and power industries. In this work, the capability of active thermography and eddy current techniques to detect real imposed defects that are representative of the laser powder bed fusion process has been investigated. A 3D complex shape of defects was revealed by a μCT investigation used as reference results for the other NDT methods. The study was focused on two different types of defects: porosities generated in keyhole mode as well as in lack of fusion mode. Different thermographic and eddy current measurements were carried out on AM samples, providing the capability to detect volumetric irregularly shaped defects using non-destructive methods.
Quality assessment of components produced by metal based additive manufacturing (AM) technologies such as laser powder bed fusion is rising in importance due to the increased use of AM in industrial production. Here, the presence of internal porosity was identified as a limiting factor for the final component quality. The utilization of thermography as an in-situ monitoring technique allows the determination of the part’s thermal history which was found to be connected to the porosity formation [1]. Combining the local thermal information derived from thermography with the porosity information obtained by x-ray micro computed tomography, machine learning algorithms can be utilized to predict the porosity distribution in the part. In this study, a first approach for the prediction of keyhole porosity in a cylindric specimen from AISI 316L stainless steel is presented. It is based on data augmentation using the “SmoteR” algorithm [2] to cure the dataset imbalance and a 1-dimensional convolutional neural network.
[1] C.S. Lough et al., Local prediction of Laser Powder Bed Fusion porosity by short-wave infrared thermal feature porosity probability maps. Journal of Materials Processing Technology, 302, p. 117473 (2022)
https://dx.doi.org/10.1016/j.imatprotec.2021.117473
[2] L. Torgo et al., SMOTE for Regression. Progress in Artificial Intelligence, Chapter 33, p. 378-289 (2013)
https://dx.doi.org/10.1007/978-3-642-40669-0_33