Refine
Year of publication
Document Type
- Article (24)
- Other (10)
- Conference Proceeding (9)
- Part of a Book (1)
- misc (1)
Keywords
- additive manufacturing (2)
- Environmental Engineering (1)
- Environmental Science (miscellaneous) (1)
- Management, Monitoring, Policy and Law (1)
- PBF-LB/M (1)
- Renewable Energy, Sustainability and the Environment (1)
- SPIT (1)
- SWIR (1)
- additive fertigung (1)
- additive manufacturing processes (1)
- bilddaten (1)
- bilderkennung (1)
- corona (1)
- covid (1)
- galvanometer scanner (1)
- in situ monitoring (1)
- internationalisierung (1)
- material (1)
- melt pool (1)
- metal printing (1)
- metalldruck (1)
- neural network (1)
- process monitoring (1)
- semantic segmentation (1)
- temperatur (1)
- thermal distortion (1)
Powder Bed Monitoring Using Semantic Image Segmentation to Detect Failures during 3D Metal Printing
(2023)
Monitoring the metal Additive Manufacturing (AM) process is an important task within the scope of quality assurance. This article presents a method to gain insights into process quality by comparing the actual and target layers. Images of the powder bed were captured and segmented using an Xception–style neural network to predict the powder and part areas. The segmentation result of every layer is compared to the reference layer regarding the area, centroids, and normalized area difference of each part. To evaluate the method, a print job with three parts was chosen where one of them broke off and another one had thermal deformations. The calculated metrics are useful for detecting if a part is damaged or for identifying thermal distortions. The method introduced by this work can be used to monitor the metal AM process for quality assurance. Due to the limited camera resolutions and inconsistent lighting conditions, the approach has some limitations, which are discussed at the end.
First time right is one major goal in powder based 3D metal printing. Reaching this goal is driven by reducing life cycle costs for quality measures, to minimize scrap rate and to increase productivity under optimal resource efficiency. Therefore, monitoring the state of the powder bed for each printed layer is state of the art in selective laser melting. In the most modern approaches the quality monitoring is done by computer vision systems working with an interference on trained neural networks with images taken after exposure and after recoating. There are two drawbacks of this monitoring method: First, the sensor signals - the image of the powder bed - give no direct height information. Second, the application of this method needs to be trained and labeled with reference images for several cases. The novel approach presented in this paper uses a laser line scanner attached to the recoating machine. With this new concept, a direct threshold measure can be applied during the recoating process to detect deviations in height level without prior knowledge. The evaluation can be done online during recoating and feedback to the controller to monitor each individual layer. Hence, in case of deviations the location in the printing plane is an inherent measurement and will be used to decide which severity of error is reported. The signal is used to control the process, either by starting the recoating process again or stopping the printing process. With this approach, the sources of error for each layer can be evaluated with deep information to evaluate the cause of the error. This allows a reduction of failure in the future, which saves material costs, reduces running time of the machine life cycle phase in serial production and results in less rework for manufactured parts. Also a shorter throughput time per print job results, which means that the employee can spent more time to other print jobs and making efficient use of the employee’s work force. In summary, this novel approach will not only reduce material costs but also operating costs and thus optimize the entire life cycle cost structure. The paper presents a first feasibility and application of the described approach for test workpieces in comparison to conventional monitoring systems on an EOS M290 machine.
Production companies are getting more and more aware of the relevancy of energy costs and the environmental impact of their manufactured products. Hence, the knowledge about the energy intensity of new process technologies as metal printing becomes increasingly crucial. Therefore, data about the energy intensity of entire process chains allow a detailed assessment of the life cycle costs and environmental impact of metal printed parts. As metal printing with Laser Powder Bed Fusion (LPBF) is applied from rapid prototyping to serial manufacturing processes more and more, sustainability data are useful to support a valid scale-up scenario and energetic improvements of the 3D-printing machinery as well as peripheral aggregates used in the process chain. The contribution aims to increase the transparency of the LPBF process chain in terms of its energy consumption. Therefore a generalized model to assess sustainability aspects of metal printed parts is derived. For this purpose, the LPBF process chain with the essential pre-, main- and post-processes is evaluated regarding its energy intensity. Here, the sub-processes, for example wet and dry cleaning of the printer, sieving of the metal powder or sand-blasting of the part are analyzed as well as the main printing process. Based on the derived experimental data from an installed, industry-like process chain, a model is created, which tends to generalize the experimental findings to evaluate other metal printed parts and process chain variants in terms of their energy intensity.
Das direkte Metall-Laser-Schmelzen (DMLS) aus der Familie der Additiven Fertigungsverfahren (AM) ermöglicht die schichtweise Erzeugung komplexer dreidimensionaler Geometrien mit hoher relativer Dichte unter Verwendung von Metallpulver als Ausgangsmaterial [1]. Die Technologie wird zunehmend eingesetzt, um innovative Bauteile material- und gewichtssparend herzustellen oder komplexe Produkte ohne zusätzliche Werkzeuge oder Spannvorrichtungen zu fertigen. Darüber hinaus sind Funktionsintegrationen, zum Beispiel Gussformen mit eingeprägten Kühlkanälen, möglich. Da einzelne Metallpulverschichten auf vorhergehende Schichten aufgeschmolzen werden, entstehen während der Herstellung des Bauteils komplexe, zeitabhängige Temperaturprofile [2]. Durch den Einsatz hoher Laserintensitäten und Scangeschwindigkeiten, bei denen die Belichtungszeit der Laserbestrahlung im Bereich von Millisekunden liegt, werden zudem extrem hohe Aufheiz- und Abkühlraten induziert, die zu einzigartigen Mikrostrukturen und Materialeigenschaften führen [3].
Diese extremen Prozessbedingungen können sich jedoch auch negativ auf den Fertigungsprozess auswirken. Bei komplexen Bauteilen bleibt die Prozessstabilität und Qualitätssicherung Umfragen zufolge weiterhin die wichtigste technologische Barriere für den Einsatz additiv gefertigter Bauteile in hochbelasteten oder sicherheitsrelevanten Bereichen [4]. Daher verspricht der Zusammenhang zwischen Temperaturprofil während der Fertigung, relativer Dichte der Bauteile, sowie thermophysikalischer Eigenschaften additiv gefertigter Proben wichtige Erkenntnisse, insbesondere im Hinblick auf eine zerstörungsfreie Qualitätssicherung, sowie neue Anwendungsmöglichkeiten.
Einsatz einer maschinell gelernten Bildsegmentierung zur Pulverbettüberwachung im Metalldruck
(2022)
Der Schwerpunkt der folgenden Ausführungen ist auf eine schichtweise Erkennung von Abweichungen durch die automatisierte Analyse von Bilddaten aus pulverbettbasierten Metalldruckprozessen gelegt. Bei diesen Prozessen wird eine dünne Schicht im Bereich von 20 bis 100 μm aus pulverförmigem Metallpulver aufgetragen. Ein zweidimensionaler Querschnitt des gewünschten Bauteils wird dann entweder mit einer selektiven Wärmequelle aufgeschmolzen oder mit einem Bindemittel zusammengebunden. Anschließend wird das Substrat um die Höhe einer Pulverschicht abgesenkt und der Vorgang wiederholt, bis der Aufbau abgeschlossen ist. Nach dem Abschluss des Aufschmelzens einer Schicht wird ein Bild mittels einer Kamera im sichtbaren Wellenlängenbereich erstellt. Abbildung 1 zeigt zwei Beispiele solcher Bilder. Diese bilden die Eingangsgröße für die Erkennung von Abweichungen. Durch die gewählte Schichtdicke kann die Herstellung eines Bauteils mehrere tausend Bilder erzeugen. Die automatisierte und zeitnahe Auswertung ist daher Inhalt aktueller Forschungs- und Entwicklungsaktivitäten [1]. Nicht zuletzt da die notwendige Sensorik – eine Kamera – wirtschaftlich und robust einsetzbar ist.
Eines der zentralen strategischen Ziele unserer Hochschule ist die Internationalisierung, sowie der »internationalisation@home«. Als die weltweite Corona-Pandemie die Präsenzlehre und -forschung ebenso wie den internationalen Austausch von Studierenden und Forschenden zu Beginn 2020 quasi zum Erliegen brachte wurden die Rufe nach digitalen Angeboten im internationalen Bereich schnell laut. Vor diesem Hintergrund reagierte der »Deutsche Akademische Auslandsdienst (DAAD)« mit der kurzfristig ins Leben gerufenen Förderlinie »International Virtual Academic Collaboration« (IVAC), um internationale Hochschulkooperationen und weltweite Mobilität unter digitalen Vorzeichen strategisch zu gestalten und auszubauen [1].
C7. 4 Application of Laser Line Scanners for Quality Control during Selective Laser Melting (SLM)
(2021)
Additive manufacturing processes, particularly Laser-Based Powder Bed Fusion of Metals (PBF-LB/M), enable the development of new application possibilities due to their manufacturing-specific freedom of design. These new fields of application require a high degree of component quality, especially in safety-relevant areas. This is currently ensured primarily via a considerable amount of downstream quality control. Suitable process monitoring systems promise to reduce this effort drastically. This paper introduces a novel monitoring method in order to gain process-specific thermal information during the manufacturing process. The Synchronized Path Infrared Thermography (SPIT) method is based on two synchronized galvanometer scanners allowing high-speed and high-resolution observations of the melt pool in the SWIR range. One scanner is used to steer the laser over the building platform, while the second scanner guides the field of view of an IR camera. With this setup, the melting process is observed at different laser powers, scan speeds and at different locations with respect to the laser position, in order to demonstrate the positioning accuracy of the system and to initially gain thermal process data of the melt pool and the heat-affected zone. Therefore, the SPIT system shows a speed independent overall accuracy of ±2 Pixel within the evaluated range. The system further allows detailed thermal observation of the melt pool and the surrounding heat-affected zone.
Nowadays, additive manufacturing processes are becoming more and more appealing due to their production-oriented design guidelines, especially with regard to topology optimisation and minimal downstream production depth in contrast to conventional technologies. However, a scientific path in the areas of quality assurance, material and microstructural properties, intrinsic thermal permeability and dependent stress parameters inhibits enthusiasm for the potential degrees of freedom of the direct metal laser melting process (DMLS). Especially in quality assurance, post-processing destructive measuring methods are still predominantly necessary in order to evaluate the components adequately. The overall objective of these investigations is to gain process knowledge make reliable in situ statements about component quality and material properties based on the process parameters used and emission values measured. The knowledge will then be used to develop non-destructive tools for the quality management of additively manufactured components. To assess the effectiveness of the research design in relation to the objectives for further investigations, this pre-study evaluates the dependencies between the process parameters, process emission during manufacturing and resulting thermal diffusivity and the relative density of samples fabricated by DMLS. Therefore, the approach deals with additively built metal samples made on an EOS M290 apparatus with varying hatch distances while simultaneously detecting the process emission. Afterwards, the relative density of the samples is determined optically, and thermal diffusivity is measured using the laser flash method. As a result of this pre-study, all interactions of the within factors are presented. The process variable hatch distance indicates a strong influence on the resulting material properties, as an increase in the hatch distance from 0.11 mm to 1 mm leads to a drop in relative density of 57.4%. The associated thermal diffusivity also reveals a sharp decrease from 5.3 mm2/s to 1.3 mm2/s with growing hatch distances. The variability of the material properties can also be observed in the measured process emissions. However, as various factors overlap in the thermal radiation signal, no clear assignment is possible within the scope of this work.
Analysis of cutting force during milling with regards to the dependency on the penetration angle
(2009)
On the way to climate neutrality manufacturing companies need to assess the Carbon dioxide (CO2) emissions of their products as a basis for emission reduction measures. The evaluate this so-called Product Carbon Footprint (PCF) life cycle analysis as a comprehensive method is applicable, but means great effort and requires interdisciplinary knowledge. Nevertheless, assumptions must still be made to assess the entire supply chain. To lower these burdens and provide a digital tool to estimate the PCF with less input parameter and data, we make use of machine learning techniques and develop an editorial framework called MINDFUL. This contribution shows its realization by providing the software architecture, underlying CO2 factors, calculations and Machine Learning approach as well as the principles of its user experience. Our tool is validated within an industrial case study.
High-temperature calibration methods in additive manufacturing involve the use of advanced techniques to accurately measure and control the temperature of the build material during the additive manufacturing process. Infrared cameras, blackbody radiation sources and non-linear optimization algorithms are used to correlate the temperature of the material with its emitted thermal radiation. This is essential for ensuring the quality and repeatability of the final product. This paper presents the calibration procedure of an imaging system for in-situ measurement of absolute temperatures and temperature gradients during powder bed fusion of metal with laser beam (PBF-LB/M) in the temperature range of 500 K–1500 K. It describes the design of the optical setup to meet specific requirements in this application area as well as the procedure for accounting the various factors influencing the temperature measurement. These include camera-specific effects such as varying spectral sensitivities of the individual pixels of the sensor as well as influences of the exposure time and the exposed sensor area. Furthermore, influences caused by the complex optical path, such as inhomogeneous transmission properties of the galvanometer scanner as well as angle-dependent transmission properties of the f-theta lens were considered. A two-step fitting algorithm based on Planck's law of radiation was applied to best represent the correlation. With the presented procedure the calibrated thermography system provides the ability to measure absolute temperatures under real process conditions with high accuracy.