TY - CONF A1 - Karapanagiotis, Christos T1 - Machine learning for strain and temperature discrimination in Brillouin distributed sensing N2 - Short presentation of the PhD project in machine learning based Brillouin distributed sensing. Machine learning can be used to enhance the performance of BOFDA and reduce considerably the measurement time. Apart from this, ML can also be used to extract more information from the Brillouin gain spectrum in order to render the temperature and strain discrimination possible T2 - PhD Day 2020 CY - Online meeting DA - 15.09.2020 KW - Fiber optics sensors KW - Machine learning PY - 2020 AN - OPUS4-51689 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Martynec, T. A1 - Karapanagiotis, Christos A1 - Klapp, S. H. L. A1 - Kowarik, S. T1 - Machine learning predictions of surface migration barriers in nucleation and non-equilibrium growth N2 - Machine learning is playing an increasing role in the discovery of new materials and may also facilitate the search for optimum growth conditions for crystals and thin films. Here, we perform kinetic Monte-Carlo simulations of sub-monolayer growth. We consider a generic homoepitaxial growth scenario that covers a wide range of conditions with different diffusion barriers (0.4–0.55 eV) and lateral binding energies (0.1–0.4 eV). These simulations are used as a training data set for a convolutional neural network that can predict diffusion barriers and binding energies. Specifically, a single Monte-Carlo image of the morphology is sufficient to determine the energy barriers with an accuracy of approximately 10 meV and the neural network is tolerant to images with noise and lower than atomic-scale resolution. We believe this new machine learning method will be useful for fundamental studies of growth kinetics and growth optimization through better knowledge of microscopic parameters. KW - Machine learning KW - Surface migration barriers KW - Material engineering PY - 2021 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-532523 UR - https://www.nature.com/articles/s43246-021-00188-1 DO - https://doi.org/10.1038/s43246-021-00188-1 SN - 2662-4443 VL - 2 SP - 1 EP - 9 PB - Springer Nature CY - London AN - OPUS4-53252 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Karapanagiotis, Christos A1 - Hicke, Konstantin A1 - Krebber, Katerina T1 - Eine Sammlung durch Maschinelles Lernen unterstützter Verteilter Faseroptischer Sensoren für Infrastrukturmonitoring T1 - A collection of machine learning assisted distributed fiber optic sensors for infrastructure monitoring N2 - In this paper, we present a collection of machine learning assisted distributed fiber optic sensors (DFOS) for applications in the field of infrastructure monitoring. We employ advanced signal processing based on artificial neural networks (ANNs) to enhance the performance of the dynamic DFOS for strain and vibration sensing. Specifically, ANNs in comparison to conventional and computationally expensive correlation and linearization algorithms, deliver lower strain errors and speed up the signal processing allowing real time strain monitoring. Furthermore, convolutional neural networks (CNNs) are used to denoise the dynamic DFOS signal and enable useable sensing lengths of up to 100 km. Applications of the machine learning assisted dynamic DFOS in road traffic and railway infrastructure monitoring are demonstrated. In the field of static DFOS, machine learning is applied to the well-known Brillouin optical frequency domain analysis (BOFDA) system. Specifically, CNN are shown to be very tolerant against noisy spectra and contribute towards significantly shorter measurement times. Furthermore, different machine learning algorithms (linear and polynomial regression, decision trees, ANNs) are applied to solve the well-known problem of cross-sensitivity in cases when temperature and humidity are measured simultaneously. The presented machine learning assisted DFOS can potentially contribute towards enhanced, cost effective and reliable monitoring of infrastructures. N2 - In diesem Beitrag stellen wir eine Sammlung von verteilten faseroptischen Sensoren (DFOS) vor, die mit Hilfe von Maschinellem Lernen arbeiten und für Anwendungen im Bereich der Infrastrukturüberwachung geeignet sind. Wir setzen hierbei fortschrittliche Signalverarbeitung auf der Grundlage Künstlicher Neuronaler Netze ein, um die Leistungsfähigkeit dynamischer DFOS für die Messung von Dehnungen und Vibrationen zu verbessern. Insbesondere Künstliche Neuronale Netze (ANNs) liefern im Vergleich zu konventionellen und rechenintensiven Korrelations- und Linearisierungsalgorithmen geringere Dehnungsfehler und beschleunigen die Signalverarbeitung, so dass eine Dehnungsüberwachung in Echtzeit möglich ist. Darüber hinaus wenden wir Convolutional Neural Networks (CNNs) an, um dynamische DFOS-Signale zu entrauschen und damit nutzbare Messlängen von bis zu 100 km zu ermöglichen. Es werden Anwendungsbeispiele dieser durch Maschinelles Lernen unterstützten dynamischen DFOS in den Bereichen des Straßenverkehrsmonitorings und der Zug- und Gleisüberwachung aufgezeigt. Im Bereich der statischen DFOS wird Maschinelles Lernen auf das Verfahren der Optischen Brillouin-Frequenzbereichsanalyse (BOFDA) angewendet. Insbesondere CNN erweisen sich hier als sehr robust gegenüber verrauschten Spektren und tragen zu deutlich kürzeren Messzeiten bei. Darüber hinaus werden verschiedene Algorithmen des maschinellen Lernens (lineare und polynome Regression, Entscheidungsbäume, ANNs) angewandt, um das bekannte Problem der Querempfindlichkeit bei DFOS in den Fällen zu lösen, in denen Temperatur und Feuchtigkeit gleichzeitig gemessen werden sollen. Die hier vorgestellten, durch Maschinelles Lernen unterstützten, DFOS können zu einer verbesserten, kostengünstigen und zuverlässigen Überwachung von Infrastrukturen beitragen. KW - Artificial neural networks KW - Brillouin sensing KW - Distributed acoustic sensing KW - Fibre optic sensors KW - Infrastructure monitoring PY - 2023 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-569044 DO - https://doi.org/10.1515/teme-2022-0098 SN - 0171-8096 SP - 1 EP - 19 PB - De Gruyter AN - OPUS4-56904 LA - mul AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - Evaluation of the generalization performance of a CNN-assisted BOFDA system N2 - Brillouin Optical Frequency Domain Analysis (BOFDA) is a powerful and well-established method for static distributed sensing of temperature and strain. Recently, we demonstrated a BOFDA system based on convolutional neural network which shortens the measurement time considerably. In this paper, we apply leave-one-out cross validation to evaluate the generalization performance and provide an unbiased and reliable machine learning model for a time-efficient BOFDA system. T2 - 21. ITG/GMA Fachtagung Sensoren und Messsysteme 2022 CY - Nuremberg, Germany DA - 10.05.2022 KW - Fiber optics sensors KW - BOFDA KW - Brillouin distributed sensing KW - Machine learning PY - 2022 UR - https://www.ama-science.org/direct/tagungsband-sensoren-und-messsysteme-2022 SN - 978-3-8007-5835-7 SP - 289 EP - 292 PB - VDE VERLAG GMBH AN - OPUS4-55183 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - Time-efficient CNN-assisted BOFDA N2 - Brillouin Optical frequency domain analysis (BOFDA) is a well-known technique in distributed Fiber sensing for temperature and strain monitoring. In this study, we applied a signal post-processing method based on convolutional neural networks (CNNs) and we show that in comparison to conventional methods based on Lorentzian curve fitting (LCF), our CNN model is more robust against noise and can shorten the measurement time by more than 9 times. This will open the way for applications, where faster monitoring is needed. T2 - Machine learning workshop CY - Online meeting DA - 18.03.2021 KW - Temperature and strain sensing KW - Machine learning KW - Distributed fiber-optic sensors PY - 2021 AN - OPUS4-52305 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Karapanagiotis, Christos A1 - Krebber, Katerina T1 - Machine Learning Approaches in Brillouin Distributed Fiber Optic Sensors N2 - This paper presents reported machine learning approaches in the field of Brillouin distributed fiber optic sensors (DFOSs). The increasing popularity of Brillouin DFOSs stems from their capability to continuously monitor temperature and strain along kilometer-long optical fibers, rendering them attractive for industrial applications, such as the structural health monitoring of large civil infrastructures and pipelines. In recent years, machine learning has been integrated into the Brillouin DFOS signal processing, resulting in fast and enhanced temperature, strain, and humidity measurements without increasing the system’s cost. Machine learning has also contributed to enhanced spatial resolution in Brillouin optical time domain analysis (BOTDA) systems and shorter measurement times in Brillouin optical frequency domain analysis (BOFDA) systems. This paper provides an overview of the applied machine learning methodologies in Brillouin DFOSs, as well as future perspectives in this area. KW - Distributed fiber optic sensors KW - BOTDA KW - BOFDA KW - Machine learning KW - Strain and temperature measurements KW - Artificial neural networks PY - 2023 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-580145 DO - https://doi.org/10.3390/s23136187 VL - 23 IS - 13 SP - 1 EP - 26 PB - MDPI AN - OPUS4-58014 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Karapanagiotis, Christos A1 - Hicke, Konstantin A1 - Krebber, Katerina T1 - Machine learning assisted BOFDA for simultaneous temperature and strain sensing in a standard optical fiber N2 - We report, to our knowledge for the first time on simultaneous distributed temperature and strain sensing in a standard telecom optical fiber using a machine learning assisted Brillouin frequency domain analysis (BOFDA) system. The well-known temperature and strain cross-sensitivity problem is addressed by developing a BOFDA system with a high signal-to-noise ratio and applying machine learning. The spectrum consists of four highly resolved peaks, whose Brillouin frequency shifts are extracted and serve as features for the machine learning algorithms. The spectra result from a 450-m standard SMF-28 optical fiber, and particularly from a segment of 30 m. This fiber segment is coiled around a stretcher and placed in a climate chamber. The applied temperature and strain values range from 20 °C to 40 °C and from 0 µɛ to 1380 µɛ, respectively. The total measurement time to achieve a high SNR and resolve four peaks with a spatial resolution of 6 m is 16 min. To discriminate temperature and strain effects, simple frequentist and more sophisticated Bayesian-based algorithms are employed with the powerful Gaussian process regression (GPR) delivering the best performance in terms of temperature and strain errors, which are found to be 2 °C and 45 µɛ, respectively. These errors are calculated using leave-one-out cross-validation, so that an unbiased estimation of the sensor’s performance is provided. KW - Machine learning KW - Brillouin distributed fiber optic sensors KW - Temperature and strain sensing KW - BOFDA PY - 2023 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-569155 DO - https://doi.org/10.1364/OE.480224 SN - 1094-4087 VL - 31 IS - 3 SP - 5027 EP - 5041 PB - OPTICA AN - OPUS4-56915 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Karapanagiotis, Christos A1 - Wosniok, Aleksander A1 - Hicke, Konstantin A1 - Krebber, Katerina T1 - Time-Efficient Convolutional Neural Network-Assisted Brillouin Optical Frequency Domain Analysis N2 - To our knowledge, this is the first report on a machine-learning-assisted Brillouin optical frequency domain analysis (BOFDA) for time-efficient temperature measurements. We propose a convolutional neural network (CNN)-based signal post-processing method that, compared to the conventional Lorentzian curve fitting approach, facilitates temperature extraction. Due to its robustness against noise, it can enhance the performance of the system. The CNN-assisted BOFDA is expected to shorten the measurement time by more than nine times and open the way for applications, where faster monitoring is essential. KW - Fiber-optic sensors KW - Machine learning KW - Temperature and strain monitoring KW - Brillouin distributed sensing PY - 2021 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-524631 DO - https://doi.org/10.3390/s21082724 VL - 21 IS - 8 SP - 2724 PB - MDPI AN - OPUS4-52463 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Karapanagiotis, Christos A1 - Hicke, Konstantin A1 - Wosniok, Aleksander A1 - Krebber, Katerina T1 - Distributed humidity fiber-optic sensor based on BOFDA using a simple machine learning approach N2 - We report, to our knowledge for the first time, on distributed relative humidity sensing in silica polyimide-coated optical fibers using Brillouin optical frequency domain analysis (BOFDA). Linear regression, which is a simple and well-interpretable algorithm in machine learning and statistics, is utilized. The algorithm is trained using as features the Brillouin frequency shifts and linewidths of the fiber’s multipeak Brillouin spectrum. To assess and improve the effectiveness of the regression algorithm, we make use of machine learning concepts to estimate the model’s uncertainties and select the features that contribute most to the model’s performance. In addition to relative humidity, the model is also able to simultaneously provide distributed temperature information addressing the well-known cross-sensitivity effects. KW - Fiber optics sensors KW - Machine learning KW - Brillouin distributed sensing KW - Humidity sensing KW - Simultaneous temperature and humidity sensing PY - 2022 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-546294 DO - https://doi.org/10.1364/OE.453906 VL - 30 IS - 8 SP - 12484 EP - 12494 PB - OPTICA AN - OPUS4-54629 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Heimann, Jan A1 - Charmi, Amir A1 - Karapanagiotis, Christos A1 - Hashemi, Seyedreza A1 - Prager, Jens T1 - How structural health monitoring can be embedded in a digital quality infrastructure: an example. N2 - The digital Quality Infrastructure (QI) initiative “QI-Digital” in Germany is focusing on implementing new technologies and approaches to ensure that the task of quality assurance is more efficient and ready for the digital and green transformation of the economy. The implementation of quality control key elements, such as Smart Standards, Digital Certificates and QI-cloud solutions shall contribute to solving the socio-economic, ecological, and technological challenges of our time. Hydrogen is a key energy carrier and has the potential to play a significant role in the energy transition, especially in mobility. An essential factor for the broad acceptance of hydrogen-based mobility is the availability of refueling stations that operate reliably and safely. Using the example of a Hydrogen Refueling Station (HRS) built within the QI-Digital initiative, the Federal Institute for Material Research and Testing (BAM) aims to establish a real laboratory where modern measurement techniques and new digital methods are implemented to enhance operational safety, availability, and economic efficiency and render the technology more attractive for the industry. In this work, we present an approach to establish a Structural Health Monitoring (SHM) system on a high-pressure buffer inside HRS and show how it could be embedded into a digital QI. The high-pressure buffers are essential components of the plant which are currently inspected periodically without regard to their operating history. Focusing on the transition to a continuous and digitally supported monitoring of the component’s integrity during operation the novel inspection scheme will be linked to a completely digitalized component-related documentation and tested using digital certificates. This allows the operational safety and, if necessary, the remaining useful lifetime to be assessed on an ongoing basis and to be a valuable contribution to increasing sustainability. T2 - 11th European Workshop on Structural Health Monitoring CY - Potsdam, Germany DA - 10.06.2024 KW - Structural health monitoring KW - Quality infrastructure KW - Digital transformation KW - Reliability KW - Quality control KW - Industry 4.0 PY - 2024 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-607109 DO - https://doi.org/10.58286/29769 SN - 1435-4934 SP - 1 EP - 7 PB - NDT.net AN - OPUS4-60710 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - Temperature and humidity discrimination in Brillouin distributed fiber optic sensing using machine learning algorithms N2 - In the last few years, the use of machine learning has emerged in the field of distributed fiber optic sensors as a promising approach to enhance their performance and provide new capabilities. In this study, we use machine learning for simultaneous measurements of temperature and humidity in polyimide (PI)-coated optical fibers based on Brillouin Brillouin optical frequency domain analysis (BOFDA). Different non-linear machine learning algorithms are employed, namely polynomial regression, decision trees and artificial neural networks (ANNs), and their discrimination performance is benchmarked against that of the conventional linear regression. The performance is evaluated using leave-one-out cross-validation to ensure that the models are reliable and able to generalize well on new data. We show that nonlinear machine learning algorithms outperform the conventional linear regression and thus could pave the way towards simultaneous cost-effective temperature and humidity distributed sensing, which has the potential to find attractive new applications in the field of civil and geotechnical engineering, from structural health monitoring of dikes and bridges to subsea cables and long pipelines corrosion detection. T2 - SPIE Photonics Europe CY - Strasbourg, France DA - 03.04.2022 KW - Distributed Brillouin sensing KW - Machine learning KW - Artificial neural networks KW - Humidity sensing KW - Simultaneous temperature and humidity sen PY - 2022 AN - OPUS4-54631 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos A1 - Hicke, Konstantin A1 - Krebber, Katerina T1 - Temperature and humidity discrimination in Brillouin distributed fiber optic sensing using machine learning algorithms N2 - In the last few years, the use of machine learning has emerged in the field of distributed fiber optic sensors as a promising approach to enhance their performance and provide new capabilities. In this study, we use machine learning for simultaneous measurements of temperature and humidity in polyimide (PI)-coated optical fibers based on Brillouin Brillouin optical frequency domain analysis (BOFDA). Different non-linear machine learning algorithms are employed, namely polynomial regression, decision trees and artificial neural networks (ANNs), and their discrimination performance is benchmarked against that of the conventional linear regression. The performance is evaluated using leave-one-out cross-validation to ensure that the models are reliable and able to generalize well on new data. We show that nonlinear machine learning algorithms outperform the conventional linear regression and thus could pave the way towards simultaneous cost-effective temperature and humidity distributed sensing, which has the potential to find attractive new applications in the field of civil and geotechnical engineering, from structural health monitoring of dikes and bridges to subsea cables and long pipelines corrosion detection. T2 - SPIE Photonics Europe 2022 CY - Strasbourg, France DA - 03.04.2022 KW - Distributed Brillouin sensing KW - Machine learning KW - Artificial neural networks KW - Simultaneous temperature and humidity sensing KW - BOFDA PY - 2022 DO - https://doi.org/10.1117/12.2620985 VL - 12139 SP - 1 EP - 7 PB - SPIE AN - OPUS4-54861 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - Evaluation of the generalization performance of a CNN-assisted BOFDA system N2 - Brillouin Optical Frequency Domain Analysis (BOFDA) is a powerful and well-established method for static distributed sensing of temperature and strain. Recently, we demonstrated a BOFDA system based on convolutional neural network which shortens the measurement time considerably. In this paper, we apply leave-one-out cross validation to evaluate the generalization performance and provide an unbiased and reliable machine learning model for a time-efficient BOFDA system. T2 - 21. ITG/GMA Fachtagung Sensoren und Messsysteme 2022 CY - Nürnberg, Germany DA - 10.05.2022 KW - Fiber optics sensors KW - BOFDA KW - Convolutional neural networks KW - Machine learning KW - Temperature sensing PY - 2022 AN - OPUS4-54862 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - Machine learning for strain, temperature and humidity discrimination in Brillouin optical frequency domain analysis N2 - We propose to our knowledge for the first time an ensemble of probabilistic machine learning models for strain, temperature and humidity discrimination in Brillouin optical frequency domain analysis (BOFDA) applying a two-fiber configuration. T2 - 28th International Conference on Optical Fiber Sensors CY - Hamamatsu, Japan DA - 20.11.2023 KW - Multiparameter sensing KW - Fiber optic sensors KW - Machine learning KW - Structural health monitoring PY - 2023 AN - OPUS4-59396 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos ED - Krebber, Katerina T1 - Machine learning for strain, temperature and humidity discrimination in Brillouin optical frequency domain analysis N2 - We propose to our knowledge for the first time an ensemble of probabilistic machine learning models for strain, temperature and humidity discrimination in Brillouin optical frequency domain analysis (BOFDA) applying a two-fiber configuration. T2 - 28th International Conference on Optical Fiber Sensors CY - Hamamatsu, Japan DA - 20.11.2023 KW - Multiparameter sensing KW - Machine learning KW - Fiber optic sensors KW - Structural health monitoring PY - 2023 SP - 1 EP - 4 PB - Optica Publishing Group AN - OPUS4-59397 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - Development of artificial neural networks for real time and high resolution strain sensing N2 - We propose a method based on artificial neural networks to extract strain information from wavelength-scanning coherent optical time domain reflectometry (C-OTDR) data. Our neural network algorithm performs more than two orders of magnitude faster than the conventional approach. This is due to the highly parallel evaluation of the neural networks on a GPU accelerated computer and the fact that conventional correlation and interpolation analysis needs many Iteration steps. This opens the way for real-time C-OTDR strain sensing because the neural Network strain predictions require less time than the measurements themselves. Real-time data Analysis enables long-term sensing e.g. in structural health monitoring, because the large amount of raw data does not have to be stored but can immediately be reduced to the strain data of interest. T2 - Workshop (in collaboration with KIT) CY - Online meeting DA - 27.08.2018 KW - Dynamic strain sensing KW - Machine learning KW - Distributed fiber-optic sensors PY - 2018 AN - OPUS4-52307 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Karapanagiotis, Christos A1 - Schukar, Marcus A1 - Krebber, Katerina T1 - Verteilte faseroptische Sensoren zur Zustandsüberwachung von Verbundwerkstoff-Druckbehältern T1 - Distributed fiber optic sensors for structural health monitoring of composite pressure vessels N2 - In diesem Beitrag geben wir einen umfassenden Überblick über unsere Forschung auf dem Gebiet der verteilten faseroptischen Sensorik für die strukturelle Zustandsüberwachung von Wasserstoffdruckbehältern aus Verbundwerkstoffen. Insbesondere zeigen wir, wie die Integration von faseroptischen Sensoren in Druckbehälter aus Verbundwerkstoffen die Sicherheit erhöht und gleichzeitig die Wartungskosten senkt. Die geringe Größe von Lichtwellenleitern ermöglicht ihre Integration in Verbundwerkstoffstrukturen während des Herstellungsprozesses, wodurch eine kontinuierliche Überwachung sowie eine präzise Erkennung und Lokalisierung von Strukturschäden während des Betriebs der Druckbehälter ermöglicht wird. Wir erörtern auch das Potenzial modernster Signalverarbeitungsmethoden und des maschinellen Lernens für die Weiterentwicklung der vorausschauenden Instandhaltung. Die von uns vorgestellten Anwendungen von faseroptischen Sensoren zeigen, dass sie einen wichtigen Beitrag zur Energiewende hin zu erneuerbaren Energien leisten können. KW - Verteilte faseroptische Sensoren KW - Verbundwerkstoffe KW - Wasserstoff-Druckbehälter KW - Strukturüberwachung KW - Maschinelles Lernen PY - 2024 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-596290 DO - https://doi.org/10.1515/teme-2023-0170 SN - 0171-8096 SP - 1 EP - 12 AN - OPUS4-59629 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - Towards predictive maintenance of hydrogen pressure vessels based on multi-sensor data fusion and digital twin modeling N2 - Hydrogen pressure vessels are among the most essential components for reliable hydrogen technology. Under current regulations, a mostly conservative strategy is employed, restricting the usage time of hydrogen pressure vessels without providing information on the real remaining lifetime. During the service life, pressure vessels are inspected periodically. However, no established method that can provide continuous monitoring or information on the remaining safe service life of the vessel. In this paper, we propose a sensor network for Structural Health Monitoring (SHM) of hydrogen pressure vessels where data from all sensors are collected and centrally evaluated. Specifically, we integrate three different SHM sensing technologies namely Guided Wave ultrasonics (GW), Acoustic Emission testing (AT), and distributed Fiber Optic Sensing (FOS). This integrated approach offers significantly more information and could therefore enable a transition from costly and time-consuming periodic inspections to more efficient and modern predictive maintenance strategies, including Artificial Intelligence (AI)-based evaluation. This does not only have a positive effect on the operational costs but enhances safety through early identification of critical conditions in the overall system in real-time. We demonstrate an experimental set-up of a lifetime test where a Type IV Composite Overwrapped Pressure Vessel (COPV) is investigated under cyclic loading instrumented with AT, FOS, and GW methods. We acquired data from the sensor network until the pressure vessel failed due to material degradation. The data collected using the three different SHM sensor technologies is planned to be evaluated individually, using data fusion, and AI. In the future, we aim to integrate the measurement setup into a hydrogen refueling station with the data stream implemented into a digital signal processing chain and a digital twin. T2 - 11th European Workshop on Structural Health Monitoring CY - Potsdam, Germany DA - 10.06.2024 KW - Hydrogen KW - Ultrasonic guided waves KW - Fiber optic sensors KW - Acoustic emission KW - Machine learning KW - Pressure vessels KW - Structural health monitoring PY - 2024 AN - OPUS4-60277 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - Structural health monitoring of hydrogen pressure vessels using distributed fiber optic sensing N2 - We report on distributed fiber optic sensing-based monitoring of hydrogen composite overwrapped pressure vessels (COPV) to simultaneously increase the operational lifespan and mitigate maintenance costs. Our approach represents, to the best of our knowledge, the first application of distributed fiber optic sensing for COPV Type IV monitoring, where the sensing fibers are attached to the surface, rather than integrated into the composite material. Specifically, we attach an optical fiber of 50 m to the pressure vessel's surface, covering both the cylindrical and dome sections. We note that our fiber optic sensing technique relies on swept wavelength interferometry providing strain information along the entire length of the optical fiber with high spatial resolution even at the millimeter scale. When the vessel is pressurized, the sensing optical fiber shows a linear strain response to pressure at every position along the fiber. After thousands of load cycles, the vessel finally fails with the optical fiber detecting and precisely localizing the damage in the vessel’s blind dome area. Furthermore, we discuss the potential of state-of-the-art signal processing methods and machine learning for advancing predictive maintenance. This could reduce the number of regular inspections, mitigate premature maintenance costs, and simultaneously increase the vessel’s remaining safe service life. We believe that the structural health monitoring of hydrogen pressure vessels with fiber optic sensors can enhance trust in hydrogen technology contributing to the energy transition in the future. T2 - 11th European Workshop on Structural Health Monitoring CY - Potsdam, Germany DA - 10.06.2024 KW - Hydrogen KW - Composites KW - Pressure vessels KW - Fiber optic sensors KW - Machine learning KW - Structural health monitoring PY - 2024 AN - OPUS4-60275 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos A1 - Schukar, Marcus A1 - Breithaupt, Mathias A1 - Krebber, Katerina T1 - Monitoring of composite pressure vessels using surface applied distributed fiber optic sensors N2 - In this paper, we report on surface-applied distributed fibre optic sensors for monitoring composite pressure vessels designed for hydrogen storage. Previous reports have revealed that integrating optical fibres within vessel composite structures effectively enables the monitoring of structural behavior throughout their lifetime. However, integrating optical fibres during the manufacturing process is complex and time-consuming. Therefore, we aim to simplify this process by attaching the optical fibres to the vessel’s surface. This method is significantly more timeefficient than the integration process and can be applied to any vessel. Our results demonstrate that surface-applied fibre optic sensors can detect and precisely localise damage. Additionally, signs of damage can be recognised even before the damage occurs. Predictive maintenance using fibre optic sensors could reduce premature maintenance costs and periodic inspections while increasing safety and extending the vessel’s useful service life. The role of machine learning in predictive maintenance is also discussed. T2 - 20th Sensors & their Applications Conference CY - Limerick, Ireland DA - 11.08.2024 KW - Hydrogen KW - Fiber optic sensors KW - Composites KW - Machine learning KW - Structural health monitoring PY - 2024 SP - 1 EP - 4 AN - OPUS4-60911 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - Monitoring hydrogen composite pressure vessels using surface applied distributed fiber optic sensors N2 - We report on surface-applied distributed fibre optic sensors for monitoring composite pressure vessels designed for hydrogen storage. Previous reports have revealed that integrating optical fibres within vessel composite structures effectively enables the monitoring of structural behavior throughout their lifetime. However, integrating optical fibres during the manufacturing process is complex and time-consuming. Therefore, we aim to simplify this process by attaching the optical fibres to the vessel’s surface. This method is significantly more timeefficient than the integration process and can be applied to any vessel. Our results demonstrate that surface-applied fibre optic sensors can detect and precisely localise damage. Additionally, signs of damage can be recognised even before the damage occurs. Predictive maintenance using fibre optic sensors could reduce premature maintenance costs and periodic inspections while increasing safety and extending the vessel’s useful service life. The role of machine learning in predictive maintenance is also discussed. T2 - 20th Sensors & their Applications Conference CY - Limerick, Ireland DA - 11.08.2024 KW - Hydrogen KW - Fiber optic sensors KW - Composites KW - Machine learning KW - Structural health monitoring PY - 2024 AN - OPUS4-60831 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - Toward a Digital Twin of Hydrogen Pressure Vessels Enabled by Distributed Fiber Optic Sensors N2 - We present a digital replica of a hydrogen pressure vessel enabled by distributed fiber optic sensors (DFOS). This digital replica dynamically displays and updates the vessel’s structural condition by calculating strain residuals defined as the difference between the measured DFOS strain and the expected strain based on pressure data. As an example, we show the ability of the DFOS to detect and localize damage caused by drilling six holes into the vessel’s body. This digital replica represents a foundational step toward a fully integrated digital twin for predictive maintenance and remaining lifetime prognosis. T2 - Sensor and Measurement Science International 2025 CY - Nuremberg, Germany DA - 06.05.2025 KW - Fiber optic sensors KW - Hydrogen KW - Digital twins KW - Structural health monitoring KW - Machine learning KW - Predictive maintenance PY - 2025 AN - OPUS4-63087 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Karapanagiotis, Christos A1 - Heimann, Jan A1 - Duffner, Eric A1 - Charmi, Amir A1 - Schukar, Marcus A1 - Hashemi, Seyedreza A1 - Prager, Jens T1 - Towards predictive maintenance of hydrogen pressure vessels based on multi-sensor data N2 - In this paper, we report on a sensor network for structural health monitoring (SHM) of Type IV composite overwrapped pressure vessels (COPVs) designed for hydrogen storage. The sensor network consists of three different SHM sensing technologies: ultrasonic guided waves (GW), acoustic emission (AE) testing, and distributed fiber optic sensors (DFOS). We present an experimental setup for a lifetime test, where a COPV is subjected to cyclic loading. Data from all sensors are collected and centrally evaluated. The COPV failed after approximately 60,000 load cycles, and the sensor network proved capable of detecting and localizing the damage even before the failure of the COPV. This multi-sensor approach offers significantly more channels of information and could therefore enable a transition from costly and time-consuming periodic inspections to more efficient and modern predictive maintenance strategies, including artificial intelligence (AI)-based evaluation. This not only has a positive effect on operational c KW - Ffiber optic sensors KW - Acoustic emission KW - Guided waves KW - Hydrogen KW - Digital twin KW - Structural health monitoring PY - 2024 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-618948 DO - https://doi.org/10.58286/30513 SP - 1 EP - 8 PB - NDT.net AN - OPUS4-61894 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos A1 - Schukar, Marcus A1 - Breithaupt, Mathias A1 - Hicke, Konstantin T1 - Toward a Digital Twin of Hydrogen Pressure Vessels Enabled by Distributed Fiber Optic Sensors N2 - We present a digital replica of a hydrogen pressure vessel enabled by distributed fiber optic sensors (DFOS). This digital replica dynamically displays and updates the vessel’s structural condition by calculating strain residuals defined as the difference between the measured DFOS strain and the expected strain based on pressure data. As an example, we show the ability of the DFOS to detect and localize damage caused by drilling six holes into the vessel’s body. This digital replica represents a foundational step toward a fully integrated digital twin for predictive maintenance and remaining lifetime prognosis. T2 - Sensor and Measurement Science International (SMSI) 2025 CY - Nuremberg, Germany DA - 06.05.2025 KW - fiber optic sensors KW - digital twin KW - structural health monitoring KW - hydrogen KW - machine learning PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-633885 UR - https://www.ama-science.org/proceedings/details/5962 SN - 978-3-910600-06-5 DO - https://doi.org/10.5162/SMSI2025/C6.1 SP - 165 EP - 166 PB - AMA Service GmbH CY - Wunstorf AN - OPUS4-63388 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos A1 - Heimann, Jan A1 - Duffner, Eric A1 - Charmi, Amir A1 - Schukar, Marcus A1 - Hashemi, Seyedreza A1 - Prager, Jens T1 - Towards predictive maintenance of hydrogen pressure vessels based on multi-sensor data fusion and digital twin modeling N2 - Hydrogen pressure vessels are among the most essential components for reliable hydrogen technology. Under current regulations, a mostly conservative strategy is employed, restricting the usage time of hydrogen pressure vessels without providing information on the real remaining lifetime. During the service life, pressure vessels are inspected periodically. However, no established method that can provide continuous monitoring or information on the remaining safe service life of the vessel. In this paper, we propose a sensor network for Structural Health Monitoring (SHM) of hydrogen pressure vessels where data from all sensors are collected and centrally evaluated. Specifically, we integrate three different SHM sensing technologies namely Guided Wave ultrasonics (GW), Acoustic Emission testing (AT), and distributed Fiber Optic Sensing (FOS). This integrated approach offers significantly more information and could therefore enable a transition from costly and time-consuming periodic inspections to more efficient and modern predictive maintenance strategies, including Artificial Intelligence (AI)-based evaluation. This does not only have a positive effect on the operational costs but enhances safety through early identification of critical conditions in the overall system in real-time. We demonstrate an experimental set-up of a lifetime test where a Type IV Composite Overwrapped Pressure Vessel (COPV) is investigated under cyclic loading instrumented with AT, FOS, and GW methods. We acquired data from the sensor network until the pressure vessel failed due to material degradation. The data collected using the three different SHM sensor technologies is planned to be evaluated individually, using data fusion, and AI. In the future, we aim to integrate the measurement setup into a hydrogen refueling station with the data stream implemented into a digital signal processing chain and a digital twin. T2 - 11th European Workshop on Structural Health Monitoring CY - Potsdam, Germany DA - 10.06.2024 KW - Acoustic emission KW - Ultrasonic guided waves KW - Fiber optic sensors KW - Hydrogen KW - Pressure vessels KW - Structural health monitoring KW - Machine learning PY - 2024 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-602764 UR - https://www.ndt.net/search/docs.php3?id=29702 SP - 1 EP - 8 PB - NDT.net AN - OPUS4-60276 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos A1 - Schukar, Marcus A1 - Breithaupt, Mathias A1 - Duffner, Eric A1 - Ulbricht, Alexander A1 - Prager, Jens A1 - Krebber, Katerina T1 - Structural health monitoring of hydrogen pressure vessels using distributed fiber optic sensing N2 - We report on distributed fiber optic sensing-based monitoring of hydrogen composite overwrapped pressure vessels (COPV) to simultaneously increase the operational lifespan and mitigate maintenance costs. Our approach represents, to the best of our knowledge, the first application of distributed fiber optic sensing for COPV Type IV monitoring, where the sensing fibers are attached to the surface, rather than integrated into the composite material. Specifically, we attach an optical fiber of 50 m to the pressure vessel's surface, covering both the cylindrical and dome sections. We note that our fiber optic sensing technique relies on swept wavelength interferometry providing strain information along the entire length of the optical fiber with high spatial resolution even at the millimeter scale. When the vessel is pressurized, the sensing optical fiber shows a linear strain response to pressure at every position along the fiber. After thousands of load cycles, the vessel finally fails with the optical fiber detecting and precisely localizing the damage in the vessel’s blind dome area. Furthermore, we discuss the potential of state-of-the-art signal processing methods and machine learning for advancing predictive maintenance. This could reduce the number of regular inspections, mitigate premature maintenance costs, and simultaneously increase the vessel’s remaining safe service life. We believe that the structural health monitoring of hydrogen pressure vessels with fiber optic sensors can enhance trust in hydrogen technology contributing to the energy transition in the future. T2 - 11th European Workshop on Structural Health Monitoring CY - Potsdam, Germany DA - 10.06.2024 KW - Hydrogen KW - Fiber optic sensors KW - Composites KW - Machine learning KW - Structural health monitoring PY - 2024 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-602731 UR - https://www.ndt.net/search/docs.php3?id=29701 SP - 1 EP - 7 PB - NDT.net AN - OPUS4-60273 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Karapanagiotis, Christos A1 - Breithaupt, Mathias A1 - Duffner, Eric A1 - Schukar, Marcus T1 - Real-time monitoring of hydrogen composite pressure vessels using surface-applied distributed fiber optic sensors N2 - In this paper, we report to the best of our knowledge for the first time on continuous real-time monitoring of composite overwrapped pressure vessels (COPVs) designed for hydrogen storage using surface-applied distributed fiber optic sensors (DFOS). We conducted continuous and real-time DFOS measurements during pressure cycling tests consisting of periodic pressure fluctuations between 20 bar and 875 bar, with a rate of 5 cycles min−1. During pressure cycling, the DFOS system measured strain changes, that under normal operating conditions were linearly correlated to changes in pressure. To detect and quantify damage-related anomalies, we trained a simple regression model to predict strain from pressure data and used the difference between predicted and measured values as a damage indicator. With our approach, the DFOS system not only detected and localized the damage but also continuously tracked its evolution in real time under dynamic pressure conditions. Furthermore, unlike previous studies where optical fibers were embedded within the composite structure, we applied them on the COPV surface, reducing both implementation cost and time while eliminating the need to modify the COPV manufacturing process. Based on our results, we are confident that DFOS can enhance safety and facilitate the transition from time-consuming periodic inspections to more efficient, machine learning-based predictive maintenance. KW - Fiber optic sensors KW - Hydrogen KW - Structural health monitoring KW - Pressure vessels KW - Predictive maintenance PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-627644 DO - https://doi.org/10.1088/2515-7647/adb9ac SN - 2515-7647 VL - 7 IS - 2 SP - 2 EP - 10 PB - IOP Publishing Ltd CY - Bristol, UK AN - OPUS4-62764 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Karapanagiotis, Christos T1 - 2. FOSsure-Projekttreffen: Ergebnisse aus den Verkehrsmonitoring-Messkampagnen T1 - 2nd FOSsure Project meeting: Results from the bridge traffic monitoring campaign N2 - We present distributed fiber optic results from the traffic monitoring measurement campaign on a newly constructed bridge in Brandenburg, which took place on 01.10.2025. Specifically, we employed a distributed acoustic sensing (DAS) system patented and developed by BAM and made use of optical fibers, which were embedded in the girders of the bridge. We show that our DAS system can capture vehicle flows along the bridge, including a) the number of vehicles, b) the number of axles, and c) estimates of vehicles' weight. T2 - 2. FOSsure Projekttreffen CY - Vienna, Germany DA - 16.10.2025 KW - Fiber optic sensors KW - Distributed acoustic sensing KW - Traffic load monitoring KW - Bridge monitoring PY - 2025 AN - OPUS4-64950 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Liehr, Sascha A1 - Jäger, L. A. A1 - Karapanagiotis, Christos A1 - Münzenberger, Sven A1 - Kowarik, Stefan T1 - Real-time dynamic strain sensing in optical fibers using artificial neural networks N2 - We propose to use artificial neural networks (ANNs) for raw measurement data interpolation and signal shift computation and to demonstrate advantages for wavelength-scanning coherent optical time domain reflectometry (WS-COTDR) and dynamic strain distribution measurement along optical fibers. The ANNs are trained with synthetic data to predict signal shifts from wavelength scans. Domain adaptation to measurement data is achieved, and standard correlation algorithms are outperformed. First and foremost, the ANN reduces the data analysis time by more than two orders of magnitude, making it possible for the first time to predict strain in real-time applications using the WS-COTDR approach. Further, strain noise and linearity of the sensor response are improved, resulting in more accurate measurements. ANNs also perform better for low signal-to-noise measurement data, for a reduced length of correlation input (i.e., extended distance range), and for coarser sampling settings (i.e., extended strain scanning range). The general applicability is demonstrated for distributed measurement of ground movement along a dark fiber in a telecom cable. The presented ANN-based techniques can be employed to improve the performance of a wide range of correlation or interpolation problems in fiber sensing data analysis and beyond. KW - Distributed vibration sensing KW - Distributed acoustic sensing KW - Coherent optical time domain reflectometry KW - Optical fiber sensor KW - Artificial neural network KW - Real time measurement KW - Distributed strain sensing PY - 2019 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-474832 UR - https://www.osapublishing.org/oe/abstract.cfm?uri=oe-27-5-7405 DO - https://doi.org/10.1364/OE.27.007405 SN - 1094-4087 VL - 27 IS - 5 SP - 7405 EP - 7425 PB - Optical Society of America AN - OPUS4-47483 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Greco, A. A1 - Starostin, V. A1 - Karapanagiotis, Christos A1 - Hinderhofer, A. A1 - Gerlach, A. A1 - Pithan, L. A1 - Liehr, Sascha A1 - Schreiber, Frank A1 - Kowarik, Stefan T1 - Fast fitting of reflectivity data of growing thin films using neural networks N2 - X-ray reflectivity (XRR) is a powerful and popular scattering technique that can give valuable insight into the growth behavior of thin films. This study Shows how a simple artificial neural network model can be used to determine the thickness, roughness and density of thin films of different organic semiconductors [diindenoperylene, copper(II) phthalocyanine and alpha-sexithiophene] on silica from their XRR data with millisecond computation time and with minimal user input or a priori knowledge. For a large experimental data set of 372 XRR curves, it is shown that a simple fully connected model can provide good results with a mean absolute percentage error of 8–18% when compared with the results obtained by a genetic least mean squares fit using the classical Parratt formalism. Furthermore, current drawbacks and prospects for improvement are discussed. KW - Artificial neural networks KW - X-ray reflectivity PY - 2019 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-498997 DO - https://doi.org/10.1107/S1600576719013311 SN - 1600-5767 VL - 52 SP - 1342 EP - 1347 PB - Wiley AN - OPUS4-49899 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -