BAM-Publica - Publikationsserver der Bundesanstalt für Materialforschung und -prüfung
Not a member yet
58839 research outputs found
Sort by
Process Monitoring with Compact NMR spectroscopy: Applications from Lab to Field
No full textThe use of compact NMR instruments based on permanent magnets has been increasing in recent years. Their affordability, portability, and ease of operation without the need for trained staff make them particularly interesting for quality control application in industrial production. Recent developments by instrument manufacturers, such as multi-nuclei options or extended interfacing, have made these systems even more versatile. However, the application of NMR spectroscopy as an online PAT tool remains rare, despite its significant potential for process optimization and control. A key challenge in exploiting this potential is the integration of lab instruments into the harsh environment of a chemical plant. Additionally, advancements in automation and data evaluation are key tasks to ensure robust, unattended operation with minimal maintenance requirements.
In this presentation, we showcase examples of using NMR spectroscopy for process monitoring at the lab scale, the development of open-source software tools for NMR data evaluation (PyIHM, within the Python package KLASSEZ), and a successful example of field integration, running an automated laboratory instrument in the environment of a industrial production plant
6. Nachtrag zum Zulassungsschein 02/BAM 4.3/03/12 für ein Kunststoff-Dränelement für Deponieoberflächenabdichtungen
No full text6. Nachtrag zum Zulassungsschein 02/BAM 4.3/03/12 für ein Kunststoff-Dränelement für Deponieoberflächenabdichtungen der Firma Freudenberg Performance Materials GmbH & Co. KG, Erlenbach am Main
Probabilistische Systemidentifikation einer Versuchsstruktur für Substrukturen von Offshore-Windenergieanlagen mit statischen und dynamischen Messdaten
No full textIn diesem Beitrag wird ein probabilistischer Ansatz zur Systemidentifikation für Tragstrukturen von Offshore-Windkraftanlagen vorgestellt. Der Schwerpunkt der Forschung liegt auf der Integration von globalen Systemantworten in Form von Eigenfrequenzen und -formen sowie Verschiebungen und Dehnungen als lokale Messdaten. Die unterschiedlichen Daten werden kombiniert für die Aktualisierung der Parameter eines Finite-Elemente-Modells genutzt. Zu diesem Zweck wird ein probabilistischer Ansatz nach Bayes verfolgt, um Vorwissen sowie Unsicherheiten einzubeziehen. Die Methodik wird bei einer Versuchsstruktur angewandt, die eine Jacket-Substruktur von Offshore-Windenergieanlagen nachbildet. Eine Systemidentifikation mit Hilfe von Überwachungsdaten ist wertvoll für Jacket-Substrukturen, da eine Zustandsanalyse für die Gewährleistung der strukturellen Integrität unerlässlich ist, aber hinsichtlich der schwierigen Offshore-Bedingungen möglichst effizient sein muss. In diesem Zusammenhang schafft diese Arbeit die Grundlage für eine Schadenserkennung, eine verbesserte Vorhersage der Ermüdungslebensdauer und optimierte Instandhaltungsstrategien. Während das Modell hinsichtlich der statischen Messdaten erfolgreich aktualisiert werden kann, sind Schwierigkeiten bei der Identifizierung der dynamischen Systemeigenschaften erkennbar
Erschütterungsprognose mit KI? Schnelle Ersatzmodelle und physikbasiertes maschinelles Lernen in der Bauwerk-Boden-Dynamik
No full textErschütterungsprognosen können mit sehr detaillierten Modellen durchgeführt werden. Dies ist sowohl bei der Erstellung des Modells (zum Beispiel für ein Finite-Element-Modell für Boden und Bauwerk), als auch bei der Berechnung zeitaufwändig, von einigen Minuten für die Wellenausbreitung in geschichteten Böden mit Wellenzahlintegralen bis zu mehreren Stunden für Randelementlösungen für die korrekte Bauwerk-Boden-Wechselwirkung. Hier sind einfache und schnelle Ersatzmodelle von Vorteil, die die Ergebnisse der detaillierten Berechnungen gut wiedergeben. Diese Ersatzmodelle können vollständig auf physikalischen Überlegungen beruhen (white-box Modelle) oder mit Hilfe von maschinellem Lernen aus einer Vielzahl von detaillierten Rechenergebnissen erzeugt werden (black-box Modelle). Erfahrungen mit black-box Modellen zeigen, dass es sinnvoll ist das maschinelle Lernen mit physikalischen Informationen anzureichern (grey-box Modelle). Es werden Anwendungsmöglichkeiten für physikbasiertes maschinelles Lernen im Bereich von Bahnerschütterungen aufgezeigt, die Erschütterungsemission durch die Fahrzeug-Fahrweg-Wechselwirkung, die Wellenausbreitung im Boden, die Erschütterungsimmission in Gebäude, Gleisschäden und das Monitoring von Eisenbahnbrücken
Laser powder bed fusion: Defect type influences critical porosity re-growth during reheating after hot isostatic pressing
No full textDespite the remarkable product design flexibility offered by additive manufacturing (AM) techniques, such as laser powder bed fusion, AM processes are susceptible to the formation of defects. In this context, the control of process parameters and the application of post-processing treatments, such as hot isostatic pressing (HIP), are of paramount importance to achieve the desired mechanical properties. The present study investigates the effectiveness of HIP as a function of process parameters in laser powder bed fused Ti-6V-4Al (PBF-LB/Ti64) using X-ray computed tomography. The process parameters are modified to obtain reference samples with low porosity, lack of fusion defects, or keyhole porosity. In all instances, subsurface keyhole porosity was observed in the as-built parts. Moreover, it was found that the efficacy of pore closure is dependent on the specific defect type. In the case of low porosity and keyhole pores, HIP resulted in effective closure. Conversely, larger lack of fusion defects were not closed due to their interconnectivity and the entrapment of argon gas. Subsequent heat treatments above the β-transus temperature allowed the investigation of the impact of defect type on porosity re-growth. For the first time, we reveal that lack of fusion defects are affected by considerable pore re-growth during post-HIP heat treatments of PBF-LB/Ti64. Such phenomenon is driven by the increasing internal pore pressure and local creep deformation at high temperatures. In contrast, re-growth is limited in samples with low porosity or keyhole pores
Assessment of in-service welding conditions for pressurized hydrogen pipelines via component test
No full textHydrogen is the energy carrier of tomorrow for a fossil-free future. This requires a reliable transport infrastructure capable of transporting large quantities of hydrogen. In addition to the construction of new pipelines, the conversion of existing natural gas (NG) networks is an essential part of global hydrogen strategies. The transport of hydrogen is fundamentally different from that of NG, as hydrogen can be absorbed into the pipeline material. Given the known effects of hydrogen embrittlement, the compatibility of the materials for the proposed pipelines (typically low alloy steels in a wide range of strengths and thicknesses) must be investigated. However, pipelines require frequent maintenance, repair, or the need to install additional outlets. In some cases, it is necessary to perform welding on or to the pipelines while they are in-service, i.e. with active gas flow under high pressure. This in-service welding poses challenges for hydrogen operations in terms of additional hydrogen absorption during welding and material compatibility. The challenge can be roughly divided into the possible austenitization of the inner pipe material exposed to hydrogen, which can lead to sufficient hydrogen absorption, and the welding itself, which causes an increased temperature range. Both lead to a significant increase in hydrogen solubility of the respective materials compared to room temperature. In this context, knowledge about welding on hydrogen pipelines is scarce due to the lack of operational experience. Fundamental experimental investigations are required to investigate the transferability from natural gas to hydrogen pipeline networks. For this reason, the present study presents a specially designed demonstrator concept for the realistic assessment of the welding process conditions. The demonstrator was designed ex-post sample extraction for quantification of the absorbed hydrogen concentration. For safety reasons, the required volume of pressurized hydrogen was limited by inserting a solid cylinder. Welding experiments on the DN50 and DN200 pressurized demonstrators showed an increased hydrogen uptake in the welded area of several ppm
New Approaches to PFAS Quali- and Quantification using GC-MS
No full textKnown as “forever chemicals”, per- and polyfluoroalkyl substances (PFAS) are a class of synthetically produced chemicals that includes an estimated 10.000 compounds. Due to the persistence, toxicity and ubiquitous occurrence, research has focused on the qualification and quantification of the most important compounds as well as on the investigation of toxicity and possible routes of entry over the last 10 years. Liquid chromatography - mass spectrometry (LC-MS) is the analytical standard to test for PFAS, as the spectrum of detectable compounds is significantly more comprehensive than it is currently the case with gas chromatography - mass spectrometry (GC-MS). However, to be able to test for PFAS contamination in a more process-independent manner and to make the analysis more widely accessible, methods based on GC-MS are currently being developed. Since GC-MS methods only cover a fraction of the compounds belonging to the PFAS group so far, further development of the corresponding measurement methods is inevitable [1, 2].
The here described work is part of the EU project 23IND13 ScreenFood [3]. The aim is to develop sensitive analytical GC-MS methods that contribute to an improved identification and quantification of various PFAS (both currently regulated and emerging PFAS) in selected food and food packaging matrices. Of particular interest as a food contact material are native and recycled polymers such as PET. Besides, various techniques, including solvent-free variants such as thermal desorption GC-MS, will be tested for a quick and easy analysis. Multiple derivatization approaches, which cover different PFAS subgroups, will also be tested and evaluated. This poster will present the overall project objectives and first results.
Acknowledgment:
The project (23IND13, ScreenFood) has received funding from the European Partnership on Metrology, co-financed from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States
Incorporating model form uncertainty in digital twins for reliable parameter updating and quantitites of interest analysis
No full textWith the rapid adoption of Digital Twins in recent years, simulation models designed to replicate real-world physical systems have become increasingly common. To achieve accurate representations, it is typically necessary to update model parameters based on observations collected from sensors or measurements of the physical asset. However, no model can fully capture the infinitely complex nature of reality. As a result, quantifying the uncertainty in model predictions is essential for reliable decision-making. Bayesian updating frameworks provide an appealing approach for parameter calibration, inherently accounting for such uncertainties. One often-overlooked source of error is model form uncertainty. This type of uncertainty arises from the fundamental discrepancies between the model and reality, stemming from the assumptions and simplifications made during model construction. Ignoring model form uncertainty can lead to overly confident predictions that fail to accurately reflect sensor observations. To address this, we propose an embedded model form uncertainty framework that attributes the model variability to a stochastic extension of the model's latent parameters. This approach enables the quantification of uncertainties that can be represented by a variation in the model parameters. Of particular interest are scenarios involving noisy observations or additional discrepancies that cannot be directly integrated into the model. By incorporating uncertainty through the parameters, this method not only quantifies uncertainty in predictions but also propagates model form uncertainty to other Quantities of Interest (QoI) that rely on the same model or its parameters. Consequently, QoI computations yield more reliable values, accounting for the potential uncertainties introduced by imperfect models during parameter updating. Moreover, this approach facilitates a more comprehensive statistical analysis of QoI distributions, offering deeper insights into the model's reliability and highlighting areas for potential improvement. By incorporating model form uncertainty, decision-makers can achieve a more robust and nuanced understanding of system behavior and prediction quality
PINNs-MPF: A Physics-Informed Neural Network framework for Multi-Phase-Field simulation of interface dynamics
No full textWe present PINNs-MPF framework, an application of Physics-Informed Neural Networks (PINNs) to handle Multi-Phase-Field (MPF) simulations of microstructure evolution. A combination of optimization techniques within PINNs and in direct relation to MPF method are extended and adapted. The numerical resolution is realized through a multi-variable time-series problem by using fully discrete resolution. Within each interval, space, time, and phases/grains are treated separately, constituting discrete subdomains. PINNs-MPF is equipped with an extended multi-networking (parallelization) concept to subdivide the simulation domain into multiple batches, with each batch associated with an independent NN trained to predict the solution. To ensure continuity across the spatio-temporal-phasic subdomains, a Master NN efficiently is to handle interactions among the multiple networks and facilitates the transfer of learning. A pyramidal training approach is proposed to the PINN community as a dual-impact method: to facilitate the initialization of training when dealing with multiple networks, and to unify the solution through an extended transfer of learning. Furthermore, a comprehensive approach is adopted to specifically focus the attention on the interfacial regions through a dynamic meshing process, significantly simplifying the tuning of hyper-parameters, serving as a key concept for addressing MPF problems using machine learning. We perform a set of systematic simulations that benchmark foundational aspects of MPF simulations, i.e., the curvature-driven dynamics of a diffuse interface, in the presence and absence of an external driving force, and the evolution and equilibrium of a triple junction. The proposed PINNs-MPF framework successfully reproduces benchmark tests with high fidelity and Mean Squared Error (MSE) loss values ranging from 10^−6 to 10^−4 compared to ground truth solutions
Transvarestraint testing of high-strength steel filler metal
No full textHigh-strength steel welds are typically not known to be susceptible to Solidification Cracking (SC). However, modern light-weight constructions may force welding in highly restrained conditions, which are known to increase the probability of Solidification Crack (SC) emergence. In this article, the Modified Varestraint-Transvarestraint (MVT) test was used to evaluate the hot cracking susceptibility of welds made from high-strength, low-alloyed filler material. The materials tested include solid wires and a metal-cored wire. All wires are typically used in the Gas Metal Arc Welding (GMAW) process. Susceptibility to SC was measured over a wide range of welding parameters and bending speeds. Results show little affinity of the tested materials to SC. However, crack length increases in most cases with arc energy ( U ∙ I∕welding speed ) and welding speed. The length of the longest crack in one test specimen follows a similar trend until high welding speeds, where stagnation of crack length with changing arc energy was observed