TY - CONF A1 - Huo, Wenjie A1 - Schmies, Lennart A1 - Gumenyuk, Andrey A1 - Rethmeier, Michael A1 - Wolter, Katinka T1 - An illumination based backdoor attack against crack detection systems in laser beam welding N2 - Deep neural networks (DNNs) have been wildly used in engineering and have achieved state-of-the-art performance in prediction and measurement tasks. A solidification crack is a serious fault during laser beam welding and it has been proven to be successfully detected using DNNs. Recently, research on the security of DNNs is receiving increasing attention because it is necessary to explore the reliability of DNNs to avoid potential security risks. The backdoor attack is a serious threat, where attackers aim to inject an inconspicuous pattern referred to as trigger into a small portion of training data, resulting in incorrect predictions in the reference phase whenever the input contains the trigger. In this work, we first generate experimental data containing actual cracks in the welding laboratory for training a crack detection model. Then, targeting this scenario, we design a new type of backdoor attack to induce the model to predict the crack as a normal state. Considering the stealthiness of the attack, a common phenomenon during the welding process, illumination, is used as the backdoor trigger. Experimental results demonstrate that the proposed method can successfully attack the crack detection system and achieve over 90% attack success rate on the test set. T2 - 8th ML4CPS 2025 – Machine Learning for Cyber-Physical Systems CY - Berlin, Germany DA - 06.03,2025 KW - System security KW - Welding crack detection KW - Backdoor attack KW - Deep neural networks PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-651357 DO - https://doi.org/10.24405/20021 VL - 2025 SP - 12 EP - 21 PB - Universitätsbibliothek der HSU/UniBw H CY - Hamburg AN - OPUS4-65135 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Neubert, Sebastian A1 - Pittner, Andreas A1 - Rethmeier, Michael T1 - Experimental determination of TRIP parameter K for mild- and high strength low alloy steels and a super martensitic filler material N2 - A combined experimental numerical approach is applied to determine the transformation induced plasticity (TRIP)-parameter K for different strength low-alloy steels of grade S355J2+N and S960QL as well as the super martensitic filler CN13-4-IG containing 13 wt% chromium and 4 wt% nickel. The thermo-physical analyses were conducted using a Gleeble® 3500 facility. The thermal histories of the specimens to be tested were extracted from corresponding simulations of a real gas metal arc weldment. In contrast to common TRIP-experiments which are based on complex specimens a simple flat specimen was utilized together with an engineering evaluation method. The evaluation method was validated with literature values for the TRIP-parameter. It could be shown that the proposed approach enables a correct description of the TRIP behavior. KW - Transformation induced plasticity (TRIP) KW - Phase transformation KW - High-strength low-alloy steel KW - Super martensitic filler material KW - Thermo physical simulation KW - Gleeble experiments PY - 2016 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-367297 DO - https://doi.org/10.1186/s40064-016-2474-0 VL - 5 IS - 575 SP - Paper 754, 1 EP - 16 PB - SpringerPlus AN - OPUS4-36729 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Tölle, Florian A1 - Gumenyuk, Andrey A1 - Gebhardt, Moritz Oliver A1 - Rethmeier, Michael T1 - Post-weld residual stress mitigation by scanning of a defocused laser beam N2 - High welding residual stresses can cause service life reducing consequences. Even though many processes have been developed to reduce these stresses, they are only applicable for wider welds and simple component geometries or are cost-intensive, respectively. The presented method uses a defocused beam after welding for heating the material regions on both sides of the weld. In this way, the welding stresses are decreased without contacting the surfaces using the available equipment. Different process parameters could be used depending on the component geometry and the laser power. The mechanism and the influence of the process parameters were investigated by FEM-simulation and experiments on S355J2+N steel and showed a stress reduction of about 73%. T2 - Lasers in Manufacturing 2011 - Proceedings of the Sixth International WLT Conference on Lasers in Manufacturing CY - Munich, Germany DA - 23.05.2011 KW - Residual stresses KW - Stress reduction KW - High energy beam welding KW - Post-weld heat treatment KW - Laser scanner optics PY - 2011 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-238482 DO - https://doi.org/10.1016/j.phpro.2011.03.052 SN - 1875-3892 VL - 12 IS - 1 SP - 410 EP - 418 PB - Elsevier CY - Amsterdam [u.a.] AN - OPUS4-23848 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Üstündag, Ö. A1 - Avilov, V. A1 - Gumenyuk, Andrey A1 - Rethmeier, Michael T1 - Full penetration hybrid laser arc welding of up to 28 mm thick S355 plates using electromagnetic weld pool support N2 - The laser hybrid welding process offers many advantages regarding deep penetration, increased welding velocity and with the help of the supplied filler wire an improved bridgeability to gap and misalignment tolerances. High power laser systems with a power of approx. 30 kW are already available on the market. Nevertheless, multi-layer technology with an arc process is still used for welding of plates from a thickness from 20 mm. A potential cause is the process instability with increasing laser power. It is inevitable that gravity drop-out due to the high hydrostatic pressure at increasing wall thickness especially at welding in flat position and with a low welding speed. The surface tension decreases with increasing root width resulting from low welding velocities. To prevent such inadmissible defects of the seam a use of weld pool support is required. Usual weld pool support systems such as ceramic or powder supports require a mechanical detachment which is time-consuming. The electromagnetic weld pool support system described in this work shows an alternative weld pool support which works contactless. It is based on generating Lorentz forces in the weld pool due to oscillating magnetic field and induced eddy currents. This innovative technology offers single pass welds up to 28 mm in flat position and reduced welding velocity with a laser power of just 19 kW. It also leads to improved mechanical-technological properties of the seams because of the slow cooling rate. With usage of an electromagnetic weld pool support the limitation of the hybrid laser arc welding process in the thick sheet metal will be extend. KW - Electromagnetic weld pool support KW - Hybrid laser arc welding KW - Thick-walled steel KW - Single pass welding PY - 2018 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-468276 DO - https://doi.org/10.1088/1742-6596/1109/1/012015 SN - 1742-6596 VL - 1109 SP - 1 EP - 7 PB - IOP Publ. AN - OPUS4-46827 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Petrat, T. A1 - Graf, B. A1 - Gumenyuk, Andrey A1 - Rethmeier, Michael ED - Schmidt, M. ED - Vollertsen, F. ED - Arnold, C. B. T1 - Laser metal deposition as repair technology for a gas turbine burner made of Inconel 718 N2 - Maintenance, repair and overhaul of components are of increasing interest for parts of high complexity and expensive manufacturing costs. In this paper a production process for laser metal deposition is presented, and used to repair a gas turbine burner of Inconel 718. Different parameters for defined track geometries were determined to attain a near net shape deposition with consistent build-up rate for changing wall thicknesses over the manufacturing process. Spot diameter, powder feed rate, welding velocity and laser power were changed as main parameters for a different track size. An optimal overlap rate for a constant layer height was used to calculate the best track size for a fitting layer width similar to the part dimension. Deviations in width and height over the whole build-up process were detected and customized build-up strategies for the 3D sequences were designed. The results show the possibility of a near net shape repair by using different track geometries with laser metal deposition. T2 - LANE - 9 International Conference on Photonic Technologies CY - Fürth, Germany DA - 19.09.2016 KW - Laser metal deposition KW - Inconel 718 KW - Additive manufacturing KW - Maintenance KW - Repair and overhaul PY - 2016 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-376723 UR - http://ac.els-cdn.com/S1875389216301857/1-s2.0-S1875389216301857-main.pdf?_tid=ed1d75de-84a2-11e6-af94-00000aab0f6c&acdnat=1474974777_4917d753cb3d316c4b000ba0760778b5 DO - https://doi.org/10.1016/j.phpro.2016.08.078 SN - 1875-3892 VL - 83 SP - 761 EP - 768 PB - Elservier AN - OPUS4-37672 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Sproesser, G. A1 - Schenker, S. A1 - Pittner, Andreas A1 - Borndörfer, R. A1 - Rethmeier, Michael A1 - Chang, Y.-J. A1 - Finkbeiner, M. T1 - Sustainable welding process selection based on weight space partitions N2 - Selecting a welding process for a given application is crucial with respect to the sustainability of part manufacturing. Unfortunately, since welding processes are evaluated by a number of criteria, preferences for one or the other process can be contradictory. However, the prevalent procedure of weight assignment for each criterion is subjective and does not provide information about the entire solution space. From the perspective of a decision maker it is important to be able to assess the entire set of possible weightings and answer the question which welding process is optimal for which set of weights. This issue is investigated by means of a weight space partitioning approach. Two welding processes are considered with respect to three criteria that reflect their economic and environmental performance. In order to find the most sustainable welding process the underlying weight space partition is evaluated. T2 - 13th Global Conference on Sustainable Manufacturing – Decoupling Growth from Resource Use CY - Bình Dương New City, Vietnam DA - 16.09.2015 KW - Welding costs KW - Multi-criteria decision support KW - LCA KW - Welding process selection KW - GMAW KW - Multi-attribute decision method PY - 2016 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-353740 DO - https://doi.org/10.1016/j.procir.2016.01.077 SN - 2212-8271 VL - 40 SP - 127 EP - 132 PB - Elsevier B.V. AN - OPUS4-35374 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Gebhardt, M.O. A1 - Gumenyuk, Andrey A1 - Rethmeier, Michael T1 - Numerical analysis of hot cracking in laser-hybrid welded tubes N2 - In welding experiments conducted on heavy wall pipes, the penetration mode (full or partial penetration) occurred to be a significant factor influencing appearance of solidification cracks. To explain the observed phenomena and support further optimization of manufacturing processes, a computational model was developed, which used a sophisticated strategy to model the material. High stresses emerged in the models in regions which showed cracking during experiments. In partial penetration welding, they were caused by the prevention of weld shrinkage due to the cold and strong material below the joint. Another identified factor having an influence on high stress localization is bulging of the weld. PY - 2013 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-302961 DO - https://doi.org/10.1155/2013/520786 SN - 1687-8442 SN - 1687-8434 IS - Article ID 520786 SP - 1 EP - 8 PB - Hindawi Publishing Corporation CY - New York, NY, USA AN - OPUS4-30296 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Bachmann, Marcel A1 - Avilov, Vjaceslav A1 - Gumenyuk, Andrey A1 - Rethmeier, Michael T1 - Experimental and numerical investigation of an electromagnetic weld pool control for laser beam welding N2 - The objective of this study was to investigate the influence of externally applied magnetic fields on the weld quality in laser beam welding. The optimization of the process parameters was performed using the results of computer simulations. Welding tests were performed with up to 20 kW laser beam power. It was shown that the AC magnet with 3 kW power supply allows for a prevention of the gravity drop-out for full penetration welding of 20 mm thick stainless steel plates. For partial penetration welding it was shown that an0.5 T DC magnetic field is enough for a suppression of convective flows in the weld pool. Partial penetration welding tests with 4 kW beam power showed that the application of AC magnetic fields can reduce weld porosity by a factor of 10 compared to the reference joints. The weld surface roughness was improved by 50%. KW - Laser beam welding KW - Electromagnetic weld pool support KW - Hartmann effect KW - Electromagnetic rectification PY - 2014 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-314405 DO - https://doi.org/10.1016/j.phpro.2014.08.006 SN - 1875-3892 VL - 56 SP - 515 EP - 524 PB - Elsevier B.V. CY - Amsterdam [u.a.] AN - OPUS4-31440 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Jonietz, Florian A1 - Ziegler, Mathias A1 - Myrach, Philipp A1 - Suwala, H. A1 - Rethmeier, Michael T1 - Untersuchung von Punktschweißverbindungen mit aktiver Thermografie N2 - Widerstandspunktschweißen ist insbesondere im Automobilbau eine der wichtigsten Fügetechniken. Bislang erfolgt die Qualitätssicherung überwiegend durch stichprobenartige zerstörende Prüfung. Eine zerstörungsfreie Prüftechnik würde neben der Reduzierung der Prüfkosten auch eine Optimierung des Punktschweißverfahrens bedeuten, da prinzipiell jeder Schweißpunkt geprüft werden könnte und somit auch eine Reduzierung der Anzahl der Schweißpunkte möglich ist. Es wird ein Verfahren vorgestellt, bei dem die Punktschweißverbindung zwischen zwei Stahlblechen optisch auf einer Blechseite mittels Laser oder Blitzlicht erwärmt wird. Die aufgeschmolzene Zone, die sogenannte Schweißlinse, stellt dabei neben der mechanischen Verbindung auch eine Wärmebrücke zwischen den beiden verschweißten Blechen dar, die bei diesem Verfahren ausgenutzt wird. Durch den verbesserten thermischen Kontakt zwischen den verschweißten Blechen an der Schweißlinse kontrastiert diese deutlich mit dem umgebenden Blechmaterial, bei dem der Wärmeübertrag zwischen den Blechen vergleichsweise gering ist. Dieser Kontrast im thermischen Verhalten kann mittels zeitabhängiger Thermografie gemessen werden. Durch das hier vorgestellte Verfahren kann mittels aktiver Thermografie sowohl in Transmissions- als auch in Reflexionsanordnung die Größe des thermischen Kontaktes zwischen den beiden Blechen ermittelt werden, welche ein Maß für die Größe der Schweißlinse und damit für die qualitative Güte der Schweißung darstellt. Ein Vorteil des entwickelten Verfahrens ist seine Anwendbarkeit auf Bleche ohne Oberflächenbehandlung. T2 - DACH-Jahrestagung 2015 CY - Salzburg, Austria DA - 11.05.2015 PY - 2015 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-343476 UR - http://www.ndt.net/?id=19068 SN - 1435-4934 VL - 21 IS - 4 SP - 1 EP - 7 PB - NDT.net CY - Kirchwald AN - OPUS4-34347 LA - deu AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CHAP A1 - Sproesser, G. A1 - Change, Y.-J. A1 - Pittner, Andreas A1 - Finkbeiner, M. A1 - Rethmeier, Michael ED - Stark, R. ED - Bonvoisin, J. ED - Seliger, G. T1 - Sustainable technologies for thick metal plate welding N2 - Welding is the most important joining technology. In the steel construction industry, e.g. production of windmill sections, welding accounts for a main part of the manufacturing costs and resource consumption. Moreover, social issues attached to welding involve working in dangerous environments. This aspect has unfortunately been neglected so far, in light of a predominant focus on economics combined with a lack of suitable assessment methods. In this chapter, exemplary welding processes are presented that reduce the environmental and social impacts of thick metal plate welding. Social and environmental Life Cycle Assessments for a thick metal plate joint are conducted for the purpose of expressing and analysing the social and environmental impacts of welding. Furthermore, it is shown that state-of-the-art technologies like Gas Metal Arc Welding with modified spray arcs and Laser Arc-Hybrid Welding serve to increase social and environmental performance in contrast to common technologies, and therefore offer great potential for sustainable manufacturing. KW - Human health G. KW - Life cycle assessment (LCA) KW - Arc welding KW - Laser arc-hybrid welding KW - Resource efficiency KW - Social life cycle assessment (SLCA) PY - 2017 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-390025 SN - 978-3-319-48513-3 SN - 978-3-319-48514-0 DO - https://doi.org/10.1007/978-3-319-48514-0 SN - 2194-0541 SN - 2194-055X SP - 71 EP - 84 PB - Springer CY - Cham, Switzerland AN - OPUS4-39002 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Yang, Chunliang A1 - Yang, Fan A1 - Meng, Xiangmeng A1 - Putra, Stephen Nugraha A1 - Bachmann, Marcel A1 - Rethmeier, Michael T1 - Phase-field simulation of the dendrite fragmentation by electromagnetic stirring in AA5754 aluminum alloy laser beam welding N2 - A phase-field model including magnetic field induced dendrite fragmentation was established and applied to the cases with different initial crystal nuclear positions for AA5754 aluminum alloy electromagnetic laser beam welding. Compare the calculated results that include dendrite fragmentation caused by the thermal electromagnetic Lorentz force with the results that consider only the thermal electromagnetic Lorentz force, without fragmentation, at the characteristic time instants. Both in the early and late stages, the small fragmentation at the dendrite tip promotes the number of higher-order branches and their growth, especially in the direction perpendicular to the solidification. The later stage fragmentation has the possibility of breaking one grain into several, which verifies the possibility of grain refinement caused by dendrite fragmentation. The fracture surface caused by fragmentation also makes more solid-liquid interfaces and their growth. In addition, the cases with different initial nuclear positions were compared. The grain growth in the low-temperature zone can be inhibited by the equiaxed grains' fragmentation at the high-temperature area (179.8 μm² and 14.7 % start at the center, 115.4 μm² and 9.4 % start at the high-temperature corner, 134.3 μm² and 10.9 % start at the low-temperature corner), which is another kind of grain refinement by the dendrite fragmentation. This kind of inhibition effect on grain growth in the low-temperature region will be enhanced with the increasing time interval between the two crystal nuclei’ appearance (179.8 μm² and 14.7 % when virtual grains appear at t = 4.3803 s and t = 4.3803 s, 134.3 μm² and 10.9 % at t = 4.0977 s and t = 3.9564 s, and 115.4 μm² and 9.4 % at t = 3.8151 s and t = 3.5325 s). KW - Laser beam welding KW - Electromagnetic KW - Aluminum alloys KW - Phase field method KW - Equiaxed grain KW - Dendrite fragmentation PY - 2026 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-656610 DO - https://doi.org/10.1016/j.cwe.2025.100014 SN - 3117-4159 VL - 35 IS - 1 SP - 1 EP - 12 PB - Elsevier B.V. AN - OPUS4-65661 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Meng, Xiangmeng A1 - Bachmann, Marcel A1 - Kising, Pascal A1 - Yang, Fan A1 - Rethmeier, Michael T1 - Prediction of weld pool and keyhole geometries in high-power laser beam welding through a physics-informed generative artificial intelligence approach N2 - The weld pool and keyhole geometries are critical characteristics in evaluating the stability of the high-power laser beam welding (LBW) process and determining the resultant weld quality. However, obtaining these data through experimental or numerical methods remains challenging due to the difficulties in experimental measurements and the high computational demands of numerical modelling. This paper presents a physics-informed generative approach for predicting weld pool and keyhole geometries in the LBW process. With the help of a well experimentally validated numerical model considering the underlying physics in the LBW, the geometries of the weld pool and keyhole under various welding conditions are calculated, serving as the dataset of the generative model. A conditional variational autoencoder model is employed to generate realistic 2D weld pool and keyhole geometries from the welding parameters. We utilize a β-variational autoencoder model with the evidence lower bound loss function and include Kullback–Leibler divergence annealing to better optimize model performance and stability during training. The generated results show a good agreement with the ground truth from the numerical simulation. The proposed approach exhibits the potential of physics-informed generative models for a rapid and accurate prediction of the weld pool geometries across a diverse range of process parameters, offering a computationally efficient alternative to full numerical simulations for process optimization and control in laser beam welding processes. KW - Laser beam welding KW - Generative artificial intelligence KW - Machine learning KW - Numerical simulation KW - Weld pool KW - Keyhole dynamics PY - 2026 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-656621 DO - https://doi.org/10.2351/7.0001862 SN - 1042-346X VL - 38 IS - 1 SP - 1 EP - 8 PB - Laser Institute of America AN - OPUS4-65662 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Fabry, Çağtay A1 - Pittner, Andreas A1 - Kaplan, Alexander A1 - Rethmeier, Michael T1 - Digitalisation and Carbon Footprint Evaluation of Gas Metal Arc Welding and Wire-Arc Directed Energy Deposition N2 - This paper showcases how a holistic approach to digitalisation enables data-driven welding applications, exemplarily for a gas metal arc welding (GMAW) laboratory. The workflow integrates advanced process monitoring, synchronised multi-sensor data acquisition and tools for data analytics. A welding domain-specific data exchange format weldx is presented that unifies and aggregates the data sets acquired during process monitoring with final component quality metrics, supporting reuse, traceability, and reproducibility. Two case studies illustrate the approach. First, GMAW parameters are adaptively adjusted according to local seam geometry to compensate joint-preparation deviations from nominal values typical for large-scale steel fabrication. Second, the seamless data aggregation along the welding production chain enables an automatic life-cycle assessment (LCA), quantifying the environmental impacts of additive manufacturing with DED Arc/M and attributing the dominant contributors to the carbon footprint. Collectively, the results indicate that a fully integrated experimental set-up together with standardised data structures and scalable analytics can couple monitoring, control, and sustainability, thereby realising the potential of digitalisation for high-quality and environmentally informed welding production. KW - Digitalisation KW - Arc welding KW - DED-arc KW - Life Cycle Assessment PY - 2026 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-656812 DO - https://doi.org/10.1088/1757-899X/1342/1/012067 SN - 1757-8981 VL - 1342 IS - 1 SP - 1 EP - 15 PB - IOP Publishing AN - OPUS4-65681 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Graf, Benjamin A1 - Gumenyuk, Andrey A1 - Rethmeier, Michael T1 - Laser metal deposition as repair technology for stainless steel and titanium alloys N2 - In a repair process chain, damaged areas or cracks can be removed by milling and subsequently be reconditioned with new material deposition. The use of laser metal deposition has been investigated for this purpose. The material has been deposited into different groove shapes, using both stainless steel and Ti-6Al-4 V. The influence of welding parameters on the microstructure and the heat affected zone has been studied. The parameters have been modified in order to achieve low heat input and consequently low distortion as well as low metallurgical impact. Finally, an evaluation of the opportunities for an automatized repair process is made. T2 - LANE 2012 - Laser Assisted net shape engineering 7 CY - Fürth, Germany DA - 12.11.2012 KW - Laser metal deposition KW - Laser powder cladding KW - Repair welding KW - Ti-6Al-4 V KW - Stainless steel PY - 2012 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-273066 DO - https://doi.org/10.1016/j.phpro.2012.10.051 N1 - Serientitel: Physics Procedia – Series title: Physics Procedia VL - 39 SP - 376 EP - 381 PB - Elsevier B.V. AN - OPUS4-27306 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Meng, Xiangmeng A1 - Bachmann, Marcel A1 - Yang, Fan A1 - Rethmeier, Michael T1 - Porosity prediction in laser beam welding with a multimodal physics-informed machine learning framework N2 - Laser beam welding (LBW) of metallic components is a knowledge‑intensive manufacturing process whose quality depends on the complex multi‑physics. However, its engineering application is often hindered by the occurrence of porosity defects. Achieving a thorough understanding and reliable prediction of porosity defects remains difficult because it demands robust representation and reasoning over nonlinear and hard‑to‑observe physical information. In this study, we propose an integrated multimodal physics-informed machine learning (PIML) framework with the help of multi-physical modelling and experimental data to predict the porosity defects in laser beam welding of aluminum alloys. The whole framework contains a multimodal PIML model for predicting the porosity ratio and an ML-based estimator for relevant physical information. By utilizing the scalar welding parameters and high-dimensional physical information (probability of keyhole collapses, cumulative existing time of collapses, and molten pool geometry) as inputs, the multimodal PIML model shows great superiority in predicting the porosity ratio, with a reduction of the mean square error by 45%, compared with the ML model trained only with welding parameters. The ML-based estimator constructed with an encoder‐decoder architecture can accurately reproduce the critical physical information within a timeframe of seconds. By integrating these two ML models, the proposed framework advances engineering informatics by offering a scalable, physics-knowledge‑centric solution for fast and accurate porosity prediction in LBW manufacturing. KW - Laser beam welding KW - Porosity defect KW - Physics-informed machine learning KW - Multimodal model KW - Simulation PY - 2026 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-657589 DO - https://doi.org/10.1016/j.aei.2026.104611 SN - 1474-0346 VL - 74 SP - 1 EP - 12 PB - Elsevier Ltd. AN - OPUS4-65758 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -