Filtern
Dokumenttyp
- Zeitschriftenartikel (11) (entfernen)
Schlagworte
- Artificial neural network (4)
- DED (3)
- Quality assurance (3)
- Additive Fertigung (2)
- Additive manufacturing (2)
- Data preparation (2)
- Electron beam welding (2)
- Process monitoring (2)
- Aluminum bronze (1)
- DED-EB (1)
Organisationseinheit der BAM
- 9 Komponentensicherheit (11) (entfernen)
Im Additive Manufacturing Verfahren Directed Energy Deposition (DED) wird bei der Verarbeitung von Werkzeugstahl in der Regel reines Argon als Schutzgas verwendet. Dabei kann die Verwendung von speziellen Schutzgasgemischen, auch bei geringen Anteilen zugemischter Gase, durchaus die Bauteilqualität positiv beeinflussen.
In Vorarbeiten der Messer SE & Co. KGaA zeigte ein gewisser Sauerstoffanteil im Schutzgas die Tendenz, den Flankenwinkel von Schweißspuren beim DED zu verbessern. In der vorliegenden Studie wurde daher detailliert untersucht in wie weit unterschiedliche Schutzgasgemische einen Einfluss auf die Qualität sowie die geometrischen Eigenschaften der additiv gefertigten Strukturen des Werkzeugstahls 1.2709 beim Laser-DED ausüben. Es erfolgten zunächst Testschweißungen in Form von Einzelspuren mit unterschiedlichen Gemischen aus dem Basisschutzgas Argon mit geringen Anteilen verschiedener Gase. Dabei wurde der Einfluss der Zusätze auf die Spurgeometrie und Aufbauqualität untersucht. Auf Basis dieser Vorversuche wurde eine Auswahl vielversprechender Gasgemische getroffen und Detailuntersuchungen in Form von Spuren, Flächen und Quadern unter Zugabe verschiedener Mengen an Zusätzen durchgeführt. Zur Bewertung des Einflusses der Schutzgasbeimengungen wurden der Flankenwinkel, die Porosität und das Gefüge der Proben anhand metallografischer Schliffe untersucht. Es zeigte sich, dass eine Zugabe von geringen Anteilen an Zusätzen zunächst zu einer Vergrößerung des Flankenwinkels im Vergleich zu reinem Argon führt. Mit steigendem Anteil der Gase nimmt dieser Winkel jedoch ab. So kann je nach Menge des zugesetzten Gases eine individuelle Benetzung des aufgetragenen Materials an der Oberfläche erreicht werden. Auch die Porosität ließ sich durch Schutzgasgemische beeinflussen und zeigt ein abweichendes Verhalten im Vergleich zu reinem Argon.
In recent years, in addition to the commonly known wire-based processes of Directed Energy Deposition using lasers, a process variant using the electron beam has also developed to industrial market maturity. The process variant offers particular potential for processing highly conductive, reflective or oxidation-prone materials. However, for industrial usage, there is a lack of comprehensive data on performance, limitations and possible applications. The present study bridges the gap using the example of the high-strength aluminum bronze CuAl8Ni6. Multi-stage test welds are used to determine the limitations of the process and to draw conclusions about the suitability of the parameters for additive manufacturing. For this purpose, optimal ranges for energy input, possible welding speeds and the scalability of the process were investigated. Finally, additive test specimens in the form of cylinders and walls are produced, and the hardness profile, microstructure and mechanical properties are investigated. It is found that the material CuAl8Ni6 can be well processed using wire electron beam additive manufacturing. The microstructure is similar to a cast structure, the hardness profile over the height of the specimens is constant, and the tensile strength
and elongation at fracture values achieved the specification of the raw material.
Directed energy deposition (DED) has been in industrial use as a coating process for many years. Modern applications include the repair of existing components and additive manufacturing. The main advantages of DED are high deposition rates and low energy input. However, the process is influenced by a variety of parameters affecting the component quality. Artificial neural networks (ANNs) offer the possibility of mapping complex processes such as DED. They can serve as a tool for predicting optimal process parameters and quality characteristics. Previous research only refers to weld beads: a transferability to additively manufactured three-dimensional components has not been investigated. In the context of this work, an ANN is generated based on 86 weld beads. Quality categories (poor, medium, and good) are chosen as target variables to combine several quality features. The applicability of this categorization compared to conventional characteristics is discussed in detail. The ANN predicts the quality category of weld beads with an average accuracy of 81.5%. Two randomly generated parameter sets predicted as “good” by the network are then used to build tracks, coatings,walls, and cubes. It is shown that ANN trained with weld beads are suitable for complex parameter predictions in a limited way.
Prognose von Qualitätsmerkmalen durch Anwendung von KI-Methoden beim Directed 10 Energy Deposition
(2022)
Dieser Beitrag enthält die Ergebnisse eines im Rahmen der DVS Forschung entwickelten Ansatzes zur Qualitätssicherung im Directed Energy Deposition. Es basiert auf der Verarbeitung verschiedener während des Prozesses gesammelter Sensordaten unter Anwendung Künstlicher Neuronale Netze (KNN). So ließen sich die Qualitätsmerkmale Härte und Dichte auf der Datenbasis von 50 additiv gefertigten Probenwürfel mit einer Abweichung < 2 % vorhersagen. Des Weiteren wurde die Übertragbarkeit des KNN auf eine Schaufelgeometrie untersucht. Auch hier ließen sich Härte und Dichte hervorragend prognostizieren (Abweichung < 1,5 %), sodass der Ansatz als validiert betrachtet werden kann.
The Directed Energy Deposition process is used in a wide range of applications including the repair, coating or modification of existing structures and the additive manufacturing of individual parts. As the process is frequently applied in the aerospace industry, the requirements for quality assurance are extremely high. Therefore, more and more sensor systems are being implemented for process monitoring. To evaluate the generated data, suitable methods must be developed. A solution, in this context, was the application of artificial neural networks (ANNs). This article demonstrates how measurement data can be used as input data for ANNs. The measurement data were generated using a pyrometer, an emission spectrometer, a camera (Charge-Coupled Device) and a laser scanner. First, a concept for the extraction of relevant features from dynamic measurement data series was presented. The developed method was then applied to generate a data set for the quality prediction of various geometries, including weld beads, coatings and cubes. The results were compared to ANNs trained with process parameters such as laser power, scan speed and powder mass flow. It was shown that the use of measurement data provides additional value. Neural networks trained with measurement data achieve significantly higher prediction accuracy, especially for more complex geometries.
The Directed Energy Deposition process is used in a wide range of applications including the repair, coating or modification of existing structures and the additive manufacturing of individual parts. As the process is frequently applied in the aerospace industry, the requirements for quality assurance are extremely high. Therefore, more and more sensor systems are being implemented for process monitoring. To evaluate the generated data, suitable methods must be developed. A solution, in this context, was the application of artificial neural networks (ANNs). This article demonstrates how measurement data can be used as input data for ANNs. The measurement data were generated using a pyrometer, an emission spectrometer, a camera (Charge-Coupled Device) and a laser scanner. First, a concept for the extraction of relevant features from dynamic measurement data series was presented. The developed method was then applied to generate a data set for the
quality prediction of various geometries, including weld beads, coatings and cubes. The results were compared to ANNs trained with process parameters such as laser power, scan speed and powder mass flow. It was shown that the use of measurement data provides additional value. Neural networks trained with measurement data achieve significantly higher prediction accuracy, especially for more complex geometries.
Effects on crack formation of additive manufactured Inconel 939 sheets during electron beam welding
(2021)
The potential of additive manufacturing for processing precipitation hardened nickel-base superalloys, such as Inconel 939 is considerable, but in order to fully exploit this potential, fusion welding capabilities for additive parts need to be explored. Currently, it is uncertain how the different properties from the additive manufacturing process will affect the weldability of materials susceptible to hot cracking. Therefore, this work investigates the possibility of joining additively manufactured nickel-based superalloys using electron beam welding. In particular,
the influence of process parameters on crack formation is investigated. In addition, hardness measurements are performed on cross-sections of the welds. It is shown that cracks at the seam head are enhanced by Welding speed and energy per unit length and correlate with the hardness of the weld metal. Cracking parallel to the weld area shows no clear dependence on the process variables that have been investigated, but is related to the hardness of the heat-affected zone.
Additive manufacturing, and therefore directed energy deposition, is
gaining more and more interest from industrial users. However, quality assurance for the components produced is still a challenge. Machine learning, especially using artificial neuronal networks, is a potential method for ensuring a high-quality standard. Based on process Parameters and monitoring data, part quality can be predicted. A further advantage is the ability to constantly learn and adopt to slight process changes.
First tests using artificial neural networks focus on the prediction of track geometry. The results show that even a small data set is enough to provide high accuracy in the predictions. In this work, an artificial neural network for the predictive analysis of relative density in laser powder cladding has been developed. A central composite experimental design is used to generate 19 data sets. Input variables are laser power, feed rate and powder mass flow. Cubes are built up where density is considered as a target value. Several neural networks are trained and evaluated with these data sets. Different topologies and initial weights are considered. The best network reaches a confidence level of around 90 % for the prediction of relative density based on the process parameters. Finally, the optimization of the generalization performance is investigated. To this purpose, methods of variation in error limit as well as cross-validation are applied. In this way, density is predictable by an artificial neural network with an accuracy of about 95 %.
Ni-based superalloys are well established in various industrial applications, because of their excellentmechanical properties and corrosion resistance at high temperatures. Despite the high development stage anda common industrial use of these alloys, hot cracking remains a major challenge limiting the weldability ofthe materials. As commonly known, the hot cracking susceptibility during welding increases with the amountof precipitation phases. Hence, a large amount of highstrength Ni-Alloys is rated as non-weldable. A newapproach based on electron beam welding at low feed rates shows great potential for reducing the hotcracking tendency of precipitation-hardened alloys. However, geometry and properties of the weld seamdiffer significantly in comparison to the common process range for practical uses. The aim of this study is toinvestigate the influence of welding parameters on the seam geometry at low feed rates between 1 mm/s and10 mm/s. For this purpose, 25 bead on plate welds on a 12 mm thick sheet made of Inconel 718 are carriedout. First, the relevant parameters are identified by performing a screening. Then the effects discovered arefurther studied by using a central composite design. The results show a significant difference between theanalyzed weld seam geometry in comparison to the well-known appearance of electron beam welded seams.