TY - CONF A1 - Ghasem Zadeh Khorasani, Media T1 - Federated SHM: a Novel Collaborative ML Approach to Overcome Data Scarcity and Data Privacy Concerns in StructuralHealth Monitoring N2 - The deployment of machine learning (ML) and deep learning (DL) in structural health monitoring (SHM) faces multiple challenges. Foremost among these is the insufficient availability of extensive high-quality data sets essential for robust training. Within SHM, high-quality data is defined by its accuracy, relevance, and fidelity in representing real-world structural scenarios (pristine as well as damaged). Although methods like data augmentation and creating synthetic data can add to datasets, they frequently sacrifice the authenticity and true representation of the data. Sharing real-world data encapsulating true structural and anomalous scenarios offers promise. However, entities are often reluctant to share raw data, given the potential extraction of sensitive information, leading to trust issues among collaborating entities. Our study introduces a novel methodology leveraging Federated Learning (FL) to navigate these challenges. Within the FL framework, models are trained in a decentralized manner across different entities, preserving data privacy. In our research, we simulated several scenarios and compared them to traditional local training methods. Employing guided wave (GW) datasets, we distributed the data among different parties (clients) using IID (independent, identically distributed or in other words, statistically identical) mini batches of dataset, as well as non-IID configurations. This approach mirrors real-world data distribution among varied entities, such as hydrogen refueling stations. In our methodology, the initial round involves individualized training for each client using their unique datasets . Subsequently, the model parameters are sent to the FL server, where they are averaged to construct a global model. In the second round, this global model is disseminated back to the clients to aid in predictive tasks. This iterative process continues for several rounds to reach convergence. Our findings distinctly highlight the advantages of FL over localized training, evidenced by a marked improvement in prediction accuracy . This research underscores the potential of FL in GW-based SHM, offering a remedy to similar challenges tied to data scarcity in other SHM approaches and paving the way for a new era of collaborative, data-centric monitoring systems. T2 - QI Digital Forum CY - Berlin, Germany DA - 10.10.2023 KW - Machine learning KW - Federated learning KW - Structural health monitoring KW - Artificial intelligence KW - Hydrogen safety KW - Data privacy PY - 2023 AN - OPUS4-59415 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Zinas, Orestis T1 - 3D Probabilistic Site Characterization N2 - The aim of the study is to infer the soil stratification from the provided CPT and borehole data. We infer the soil type at any location within the domain of interest from the SBT index Ic (Robertson, 2009). This index can be directly related to the CPT data through an empirical correlation model. In addition, the soil classes contained in the borehole logs can be expressed as bounds on Ic. A log-transformation was applied to Ic, Y = ln(Ic), and Y was modelled by a 3D Random Field, with a fully Bayesian hierarchical Gaussian Process model to explicitly capture uncertainties. T2 - 19th eawe PhD Seminar CY - Hannover, Germany DA - 06.09.2023 KW - Wind Energy KW - Site-characterization KW - Probabilistic PY - 2023 AN - OPUS4-58940 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Üstündag, Ömer T1 - Process advantages of laser hybrid welding compared to conventional arc-based welding processes for joining thick steel structures of wind tower N2 - The most common welding processes when joining thick-walled steels in the industry are arc-based welding processes such as GMAW or SAW. For this purpose, the sheets are joined in multi-layer technique, which can lead to productivity losses due to high welding times. The process-specific challenges in welding thick steels using multi-layer technique relate to the high heat input from the process. Therefore, alternative welding processes are being actively sought. A suitable alternative is provided by beam-based welding processes such as the laser beam or laser hybrid welding processes, which are characterized by deep penetration welds and lower heat input. With implementation of the laser hybrid welding process in the heavy industry, such as the wind tower industry, economic benefits can be reached such as the increase in productivity by reducing the layer number, and the lower consumption of filler material and energy. When comparing SAW welded 25 mm thick steels in five to six layers and single-pass laser hybrid welding, the welding time can be reduced more than 80 % and the costs of filler material, flux and energy can be saved up to 90 %. However, the industrial use of the laser hybrid welding process is still limited to applications, where the material thickness does not exceed 15 mm due to some process-specific challenges such as the sagging, sensitivity to manufacturing tolerances such as gaps and misalignment, limited filler wire mixing, and deteriorated mechanical properties resulting from high cooling rates. To overcome these challenges, an electromagnetic backing was used within the framework of this work, which works contactless and is based on an externally applied AC magnetic field. Eddy currents are induced due to the oscillating magnetic field, and an upward-oriented Lorentz force is generated to counteract the droplets formed due to gravitational forces. It allows to weld up to 30 mm thick structural steels in a single-pass with a 20-kW fiber laser system. Additionally, the gap bridgeability and the misalignment of edges were increased to 2 mm when welding 20 mm thick steels. With the aid of the AC magnetic field, a vortex was formed in the weld root, which had a positive effect on the filler wire mixing. A further significant advantage of the EM backing was the possibility to expand the process parameter window to maintain desired cooling times and mechanical properties, without suffering adverse effects concerning the root quality of the weld. T2 - 19th Nordic Laser Material Processing Conference CY - Turku, Finland DA - 22.08.2023 KW - Laser hybrid welding KW - Wind energy PY - 2023 AN - OPUS4-58616 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - el Dsoki, Chalid T1 - Anforderungen für den Transport kritisch-defekter Lithiumbatterien nach Sondervorschrift 376 N2 - Die BAM legt als zuständige Behörde die zusätzlichen Prüfanforderungen an die verwendeten Verpackungen/ Großverpackungen nach P 911 und LP 906 fest (Teil A dieser BAM-GGR). Außerdem kann die BAM gemäß Sondervorschrift 376 davon abweichende alternative Verpackungs- und/oder Beförderungsbedingungen zulassen (Einzelfallfestlegung,Teil B dieser BAM-GGR). T2 - 19. Gefahrgut Technik Tage CY - Berlin, Germany DA - 16.11.2023 KW - Gefahrgutregel Transport kritische Batterien PY - 2023 AN - OPUS4-59067 LA - deu AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - el Dsoki, Chalid T1 - Hitzebeständigkeit von Funkgeräten N2 - Lithium-Akkus können bei großer Hitzeeinwirkung thermisch durchgehen. Und doch sind sie in vielen Geräten der Feuerwehr verbaut, zum Beispiel in Funkgeräten. Ein Sicherheitsrisiko beim Löscheinsatz? KW - Funkgeräte Löscheinsatz PY - 2023 IS - 12 SP - 64 EP - 66 AN - OPUS4-59075 LA - deu AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Charmi, Amir A1 - Mustapha, Samir A1 - Yilmaz, Bengisu A1 - Heimann, Jan A1 - Prager, Jens T1 - A Machine Learning Based-Guided Wave Approach for Damage Detection and Assessment in Composite Overwrapped Pressure Vessels N2 - The applications of composite overwrapped pressure vessels (COPVs) in extreme conditions, such as storing hydrogen gases at very high pressure, impose new requirements related to the system's integrity and safety. The development of a structural health monitoring (SHM) system that allows for continuous monitoring of the COPVs provides rich information about the structural integrity of the component. Furthermore, the collected data can be used for different purposes such as increasing the periodic inspection intervals, providing a remaining lifetime prognosis, and also ensuring optimal operating conditions. Ultimately this information can be complementary to the development of the envisioned digital twin of the monitored COPVs. Guided waves (GWs) are preferred to be used in continuous SHM given their ability to travel in complex structures for long distances. However, obtained GW signals are complex and require advanced processing techniques. Machine learning (ML) is increasingly utilized as the main part of the processing pipeline to automatically detect anomalies in the system's integrity. Hence, in this study, we are scrutinizing the potential of using ML to provide continuous monitoring of COPVs based on ultrasonic GW data. Data is collected from a network of sensors consisting of fifteen Piezoelectric (PZT) wafers that were surface mounted on the COPV. Two ML algorithms are used in the automated evaluation procedure (i) a long short-term memory (LSTM) autoencoder for anomaly detection (defects/impact), and (ii) a convolutional neural network (CNN) model for feature extraction and classification of the artificial damage sizes and locations. Additional data augmentation steps are introduced such as modification and addition of random noise to original signals to enhance the model's robustness to uncertainties. Overall, it was shown that the ML algorithms used were able to detect and classify the simulated damage with high accuracy. T2 - 13th European Conference on Non-Destructive Testing (ECNDT) 2023 CY - Lisbon, Portugal DA - 03.07.2023 KW - Machine learning KW - Structural health monitoring KW - COPV KW - Guided waves KW - Damage localization PY - 2023 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-590789 DO - https://doi.org/10.58286/28079 SN - 2941-4989 VL - 1 IS - 1 SP - 1 EP - 6 PB - NDT.net CY - Mayen AN - OPUS4-59078 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Mustapha, S. A1 - Yilmaz, Bengisu A1 - Heimann, Jan A1 - Brence, Blaz A1 - Prager, Jens T1 - Guided Waves Propagation in Composite Overwrapped Pressure Vessel N2 - The application of composite overwrapped pressure vessels (COPV) to store hydrogen and other compressed gases, especially when operating at high pressures, imposed the need for an innovative and reliable approach to ensure the safe operation of the system. Continuous structural health monitoring (SHM) based on ultrasonic guided waves (GWs) is a promising approach due to the ability of the wave to propagate for long distances and go around complex structures, moreover the high sensitivity to various failure modes such as delamination, matrix cracking and debonding. In this study, we scrutinize the behavior of (GWs) within the COPV using a network of PZT elements that are used for excitation and sensing. A laser doppler vibrometer (LDV) was also used to scan the surface of the vessel in various directions. L(0,1) and L(0,2) were observed in the captured signal. The L(0, 2) appeared to be a dominant mode in the COPV and was capable of propagating along the entire length and maintaining a good signal-to-noise ratio. The L(0,2) mode maintained the same phase velocity when it is captured at various excitation angles (0, 45, and 90 degrees). The reduced effect of the orthotropy of the materials on the propagating waves is an important result as it will reduce the complexity in data processing when performing damage identification. T2 - 13th. European Conference on Non-Destructive Testing (ECNDT) CY - Lisbon, Portugal DA - 03.07.2023 KW - Composite Overwrapped Pressure Vessels KW - Guided Waves KW - Laser Doppler Vibrometer KW - Structural Health Monitoring KW - Sensors Placement PY - 2023 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-591081 DO - https://doi.org/10.58286/28068 SN - 2941-4989 VL - 1 IS - 1 SP - 1 EP - 6 PB - NDT.net CY - Mayen AN - OPUS4-59108 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Yilmaz, Bengisu A1 - Heimann, Jan A1 - Mustapha, S. A1 - Charmi, Amir A1 - Prager, Jens T1 - Guided wave ultrasonic feature determination in Type IV composite overwrapped pressure vessels towards the digital twin N2 - The digitalization of quality control processes and the underlying data infrastructures for safety relevant components, such as hydrogen pressure vessels, plays a significant role in the transition towards Industry 4.0. In the current safety regulations for hydrogen pressure vessels, there is no established concept for structural health monitoring. The development of a reliable structural health monitoring methodology for monitoring the structural integrity of pressure vessels enables a fast-forward transition from personnel- and costintensive recurring inspections, a.k.a. periodic maintenance, to predictive maintenance. In the work presented; we investigated the application of ultrasonic guided wave propagation to monitor and assess the condition of Type IV composite overwrapped pressure vessel (COPV). A sensor network of fifteen piezo-electric wafers is placed on the carbon fibre reinforced composite cylinder. Five different artificial damage configurations are created by gluing two different weight blocks on three different locations. The database containing measured guided wave data sets is enriched by two different boundary conditions. We utilized an open-source software, openBIS labnotebook, to store and analyse experimental datasets. The guided wave ultrasonic signals were investigated and analysed by using commonly used ultrasonic features (e.g., amplitude, frequency, time of flight) as well as non-traditional time-series features (kurtosis, skewness, variance). The features were used to calculate damage index and the detection performance for the results has been evaluated. The results suggest that both traditional and non-traditional features assume significant importance in artificial damage detection. The future works will additionally involve the impacts of operational conditions, such as periodic pressure variations temperature loadings as well as material degradations. T2 - 13th European Conference on Non-Destructive Testing (ECNDT) CY - Lisbon, Portugal DA - 03.07.2023 KW - Industry 4.0 KW - Composite overwrapped pressure vessels KW - Structural health monitoring KW - Ultrasonic guided waves KW - Predictive maintenance PY - 2023 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-591094 DO - https://doi.org/10.58286/28072 SN - 2941-4989 VL - 1 IS - 1 SP - 1 EP - 6 PB - NDT.net CY - Mayen AN - OPUS4-59109 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Ritz, Sebastian A1 - Loewe, Anna A1 - Bauer, Jan T1 - Specialties of HAZID-Study for Large Unmanned Underwater Vehicles N2 - This paper scopes the specialties of a hazard identification study for large and extra-large unmanned undersea vehicles (UUVs). A generic node/function structure is derived from different large and extra-large UUV designs, partially own vehicle designs from research projects, and partially from commercial vehicles. For each node, a short overview of its components and the proper function or operations is defined. A set of guide words is used to prompt a hazard discussion for each node, which identifies unwanted functions, resulting in potential hazards and unmitigated consequences. Related to the SafeMASS-Report from DNV, this work concentrates on selected topics of the bridge-related function on voyage, control & monitoring, and abnormal situations. For unmanned vehicles without any person on board, these three topics become more important for underwater vehicles with very limited connectivity or temporary disruptions of communication whereby an operator in a remote-control center could only be provided with rough vehicle or mission states. Therefore, vehicle control, whether in hard- or software, must handle most tasks belonging to the considered functions automatically or autonomously. Hence, the most appropriate reactions of the system to unwanted system behavior must be implemented. This HAZID study could, on the one hand, be used as background for such an implementation, and on the other hand, it is used to verify the implemented actions on the risks and hazards. T2 - OCEANS 2023 - Limerick CY - Limerick, Ireland DA - 05.06.2023 KW - Autonomous underwater vehicles KW - HAZID KW - Risk-based design KW - Safety KW - Cyber security PY - 2023 DO - https://doi.org/10.1109/OCEANSLimerick52467.2023.10244358 SP - 1 EP - 10 PB - IEEE AN - OPUS4-59099 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Tschirschwitz, Rico T1 - BEV Accidents - Consequences in Case of a Battery Thermal Runaway N2 - The number of cars which use a battery for propulsion, like BEV (battery electric vehicle) or PHEV (plug-in hybrid electric vehicle), is increasing. Following, the probability of these car types being involved in car accidents will increase. In a worst-case accident scenario, the installed batteries could go under thermal runaway (TR). A TR could be initiated by various causes, e.g., high temperatures from outside, mechanical damage of the cells/modules/batteries or internal and external short circuits. During such an event, large amounts of heat and toxic and/or flammable gases are released. This poses a great hazard to people and property in the immediate vicinity. The paper describes a large-scale test series on the lithium-ion battery TR consequences of automobile cells and modules up to E = 6.85 kWh. Near field temperatures of ΔT > 800°C in a distance of l = 2 m were measured. During an overcharge test, fragment throwing distances of l > 30 m were detected. Subsequent gas explosions of the released gases were documented. Hydrogen fluoride (HF) was measured in all tests, from cell to module. The highest measured concentration was cHF = 76 ppm, which is significantly higher than the 30-min-level for AEGL 2 (acute exposure guideline level 2). Based on the experimentally determined consequences, concrete measures for the accident analysis are subsequently drawn. T2 - 31st Annual Congress of the European Association for Accident Research and Analysis (EVU) CY - Limassol, Cyprus DA - 05.10.2023 KW - Consequences KW - Electrical Energy Storage KW - Lithium-ion battery KW - BEV KW - Gas release PY - 2023 AN - OPUS4-58547 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -