TY - JOUR A1 - Oster, Simon A1 - Scheuschner, Nils A1 - Chand, Keerthana A1 - Altenburg, Simon A1 - Gerlach, G. T1 - Potentials and challenges of deep-learning-assisted porosity prediction based on thermographic in-situ monitoring in laser powder bed fusion N2 - 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. T2 - XXXVII. Messtechnisches Symposium 2023 CY - Freiburg, Germany DA - 27.09.2023 KW - Porosity prediction KW - Defect detection KW - Laser powder bed fusion (PBF-LB/M, L-PBF) KW - Selective laser melting KW - Thermography KW - Machine learning PY - 2023 DO - https://doi.org/10.1515/teme-2023-0062 SN - 0171-8096 SN - 2196-7113 VL - 90 SP - 85 EP - 96 PB - De Gruyter CY - Berlin AN - OPUS4-58366 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Pelkner, Matthias A1 - Stegemann, Robert A1 - Sonntag, Nadja A1 - Pohl, Rainer A1 - Kreutzbruck, Marc T1 - Benefits of GMR sensors for high spatial resolution NDT applications N2 - Magneto resistance sensors like GMR (giant magneto resistance) or TMR (tunnel magneto resistance) are widely used in industrial applications, examples are position measurement and read heads of hard disk drives. However, in case of non-destructive testing (NDT) applications these sensors, although their properties are outstanding like high spatial resolution, high field sensitivity, low cost and low energy consumption, never reached a technical transfer to an application beyond sci-entific scope. This paper deals with benefits of GMR/TMR sensors in terms of high spatial resolution testing for different NDT applica-tions. The first example demonstrates the preeminent advantages of MR-elements compared with conventional coils used in eddy current testing (ET). The probe comprises one-wire excitation with an array of MR elements. This led to a better spatial resolution in terms of neighboring defects. The second section concentrates on MFL-testing (magnetic flux leakage) with active field excitation during and before test-ing. The latter illustrated the capability of highly resolved crack detection of a crossed notch. This example is best suited to show the ability of tiny magnetic field sensors for magnetic material characterization of a sample surface. Another example is based on characterization of samples after tensile test. Here, no external field is applied. The magnetization is only changed due to external load and magnetostriction leading to a field signature which GMR sensors can resolve. This gives access to internal changes of the magnetization state of the sample under test. T2 - QNDE 2017 CY - Provo, Utah, USA DA - 16.07.2017 KW - GMR KW - Non-destructive testing KW - Sensor arrays KW - Spatial resolution PY - 2018 SN - 978-0-7354-1644-4 DO - https://doi.org/10.1063/1.5031535 SN - 0094-243X VL - 1949 SP - UNSP 040001, 1 EP - 10 AN - OPUS4-45050 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Oster, Simon A1 - Scheuschner, Nils A1 - Chand, Keerthana A1 - Altenburg, Simon A1 - Gerlach, Gerald T1 - Potentials and challenges of deep-learning-assisted porosity prediction based on thermographic in-situ monitoring in PBF-LB/M N2 - 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. T2 - XXXVII. Messtechnisches Symposium 2023 CY - Freiburg im Breisgau, Germany DA - 27.09.2023 KW - Porosity prediction KW - Defect detection KW - Laser Powder Bed Fusion (PBF-LB/M, L-PBF) KW - Selective Laser Melting KW - Thermography KW - Machine Learning PY - 2023 AN - OPUS4-59192 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Oster, Simon A1 - Scheuschner, Nils A1 - Chand, Keerthana A1 - Breese, Philipp Peter A1 - Becker, Tina A1 - Heinrichsdorff, F. A1 - Altenburg, Simon T1 - From Thermographic In-situ Monitoring to Porosity Detection – A Deep Learning Framework for Quality Control in Laser Powder Bed Fusion N2 - 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. T2 - Sensor and Measurement Science International CY - Nurnberg, Germany DA - 08.05.2023 KW - Laser powder bed fusion KW - In-situ monitoring KW - Thermography KW - Machine Learning KW - Porosity PY - 2023 AN - OPUS4-57614 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -