Analytische Chemie
Filtern
Dokumenttyp
- Vortrag (12)
- Zeitschriftenartikel (2)
- Posterpräsentation (2)
- Beitrag zu einem Tagungsband (1)
Sprache
- Englisch (14)
- Deutsch (2)
- Mehrsprachig (1)
Schlagworte
- Machine Learning (17) (entfernen)
Organisationseinheit der BAM
- 8 Zerstörungsfreie Prüfung (13)
- 8.0 Abteilungsleitung und andere (9)
- 8.5 Röntgenbildgebung (5)
- 6 Materialchemie (4)
- 6.3 Strukturanalytik (4)
- 8.4 Akustische und elektromagnetische Verfahren (1)
- 8.6 Faseroptische Sensorik (1)
- 9 Komponentensicherheit (1)
- 9.6 Additive Fertigung metallischer Komponenten (1)
Eingeladener Vortrag
- nein (12)
Wir laden zum Trainingsworkshop Datenanalyse ein. Angetrieben durch die Digitalisierung und die sogenannte Industrie 4.0 steigt die Erwartungshaltung gegenüber der Nutzung von Daten. Die Unsicherheit, Widersprüchlichkeit und Redundanz separat verarbeiteter einzelner Quellen soll durch die synergetische Zusammenführung heterogener Datensätze überwunden werden. Dabei steht der zunehmenden Aufgabenkomplexität eine ebenso zunehmende Verfügbarkeit an Daten und Analyseverfahren gegenüber.
Der Workshop vermittelt ein konzeptionelles Verständnis für moderne Datenanalyseverfahren (Machine Learning (ML), Multivariate Statistik) und soll durch ein anschließendes Hands-On Training mit Python (https://www.python.org) einen einfachen Einstieg in die Thematik ermöglichen. Der Kurs richtet sich an den wissenschaftlichen Nachwuchs.
We have arrived in the data age. But why is it so difficult for the NDT community to achieve real breakthroughs with data-driven science? In this seminar, we will take a brief look at the evolution of mainstream data science to understand why the most exciting times are perhaps just ahead. We will give an overview of our activities in the junior research group 8.K which are aimed at enabling the next generation of data science methods in NDT. The seminar addresses the two main work fields of our group: semantic data management and the handling of limited data resources.
The first field addresses the problem that a uniform representation of our data is not yet available. However, knowledge creation in data science - whose main contribution lies in the analysis of distributed resources - requires common data access based on a collective understanding. To achieve this, we present an ontology-based approach. Ontologies are already the core of many intelligent systems such as building information models or research databases. We summarize some of the basic principles of this technology and describe our approach to create an NDT ontology.
The second field ties in with the first and addresses the application of data-based methods in engineering practice. Especially in the field of non-destructive testing many successful applications have been published. In most cases, however, the creation of referenced data is extremely expensive and therefore much sparser than in other research areas. As a result, the available data may cover only one scenario, so that common benchmarks often do not reflect the actual performance of the model in practical applications. Estimates that quantify the transferability from one scenario to another are not only necessary to overcome this challenge - they also prove to be a powerful tool for the strategic expansion of what we consider knowledge.
Die Verwendung von Methoden des Maschinellen Lernens (ML) und der Künstlichen Intelligenz (KI) im Fachbereich 8.6 Faseroptische Sensorik wird dargestellt. Die vielfältigen Möglichkeiten, Machine Learning auf Basis Künstlicher Neuronaler Netze (ANN) für eine schnelle und effiziente Datenverarbeitung eizusetzen werden demonstriert. Hierfür werden Beispiele für die Anwendungszwecke Messgrößenberechnung, Entrauschen, Interpolation, Bildverarbeitung und Messdatenauswertung aufgezeigt.
Machine learning based multi-sensor fusion for the nondestructive testing of corrosion in concrete
(2020)
Half-cell potential mapping (HP) is the most popular non-destructive testing method (NDT) for locating corrosion damage in concrete. It is generally accepted that HP is susceptible to environmental factors caused by salt-related deterioration, such as different moisture and chloride gradients. Additional NDT methods are able to identify distinctive areas but are not yet used to estimate more accurate test results. We present a Supervised Machine Learning (SML) based approach to data fusion of seven different signal features to obtain higher quality information. SMLs are methods that explore (or learn) relationships between different (sensor) data from predefined data labels. To obtain a representative, labelled data set we conducted a comprehensive experiment simulating the deterioration cycle of a chloride exposed device in the laboratory. Our data set consists of 18 measurement campaigns, each containing HP, Ground Penetrating- Radar, Microwave Moisture and Wenner resistivity data. We compare the performance of different ML approaches. Many outperform the best single method, HP. We describe the intrinsic challenges posed by a data-driven approach in NDT and show how future work can help overcome them.
Das maschinelle Lernen (ML) wurde erfolgreich zur Lösung vieler Aufgaben in der zerstörungsfreien Prüfung im Bauwesen (ZfPBau) eingesetzt. Allerdings ist die Erstellung von Referenzdaten in den meisten Fällen extrem teuer und daher viel knapper als in anderen Forschungsbereichen. Auch decken die verfügbaren Daten mitunter nur ein einziges Szenario ab, so dass die Leistungsindikatoren oft nicht die tatsächliche Leistung des ML-Modells in der praktischen Anwendung widerspiegeln. Schätzungen, die die Übertragbarkeit von einem Szenario auf ein anderes quantifizieren, sind erforderlich, um dieser Herausforderung gerecht zu werden und den Weg für Anwendungen in der Praxis zu ebnen.
In diesem Beitrag stellen wir Werkzeuge zur Beschreibung der Unsicherheit von ML in neuen ZfPBau-Szenarien vor. Zu diesem Zweck haben wir einen bestehenden Trainingsdatensatz zur Klassifizierung von Korrosionsschäden der Bewehrung in Beton um eine neue Fallstudie erweitert. Die Messungen wurden an großflächigen Betonproben mit eingebauter chloridinduzierter Korrosion des Bewehrungsstahls durchgeführt. Das Experiment simulierte den gesamten Lebenszyklus von chloridinduzierten Sichtbetonbauteilen im Labor. Unser Datensatz umfasst Potenzialfeld- und Radarmessungen. Die einzigartige Fähigkeit, die Schädigung zu überwachen und eine gezielte Korrosion einzuleiten, ermöglichte es, die Daten zu labeln - was für die Konstruktion von ML-Modellen entscheidend ist. Um die Übertragbarkeit zu untersuchen, erweitern wir unser Modell um Metadaten - wie etwa Konstruktionsmerkmale des Prüfkörpers und Umweltbedingungen. Dies erlaubt es, die Veränderung dieser Merkmale in neuen Szenarien mit statistischen Methoden als Unsicherheiten auszudrücken. Wir vergleichen verschiedene auf Stichproben und statistischer Verteilung basierende Ansätze und zeigen, wie diese Methoden eingesetzt werden können, um Wissenslücken von ML-Modellen in der ZfP zu schließen
Understanding on how a machine learning model interprets data is a crucial step to verify its reliability and avoid overfitting. While the focus of the scientific community is nowadays orientated towards deep learning approaches, which are considered as black box approaches, this work presents a toolbox that is based on complementary methods of feature extraction and selection, where the classification decisions of the model are transparent and can be physically interpreted. On the example of guided wave benchmark data from the open guided waves platform, where delamination defects were simulated at multiple positions on a carbon fiber reinforced plastic plate under varying temperature conditions, the authors could identify suitable frequencies for further investigations and experiments. Furthermore, the authors presented a realistic validation scenario which ensures that the machine learning model learns global damage characteristics rather than position specific characteristics.
To date, the destructive extraction and analysis of drilling cores is the main possibility to obtain depth information about damaging water ingress in building floors. The time- and costintensive procedure constitutes an additional burden for building insurances that already list piped water damage as their largest item. With its high sensitivity for water, a ground-penetrating radar (GPR) could provide important support to approach this problem in a non-destructive way. In this research, we study the influence of moisture damage on GPR signals at different floor constructions. For this purpose, a modular specimen with interchangeable layers is developed to vary the screed and insulation material, as well as the respective layer thickness. The obtained data set is then used to investigate suitable signal features to classify three scenarios: dry, damaged insulation, and damaged screed. It was found that analyzing statistical distributions of A-scan features inside one B-scan allows for accurate classification on unknown floor constructions. Combining the features with multivariate data analysis and machine learning was the key to achieve satisfying results. The developed method provides a basis for upcoming validations on real damage cases.
Irrespective of the experimental care used to acquire Computed Tomography Data, certain artifacts might still exist such as: Noise, Blurring, Ring Artifacts etc. To tackle this problem, a complete multi-level framework employing AI (Deep Artificial Neural Nets), targeting specific artifacts individually, is presented. The goal is to render the data suitable for subsequent unproblematic segmentation without any loss of information, compared to manual conditioning with traditional filters.
The strategy can therefore be used to acquire faster CT data (e.g. in-situ investigations) and ensure legacy with existing data obtained, perhaps, on older instruments.
In this talk an overview about artificial intelligence/machine learning applications @BAMline is given. In the first part, the use of neural networks for the quantification of XRF measurements and the decoding of coded-aperture measurements are shown. Then it is shown how Gaussian processes and Bayesian statistics can be used to achieve an optimal alignment of the set-up and in general for optimization of measurements.
Getting more efficient – The use of Bayesian optimization and Gaussian processes at the BAMline
(2022)
For more than 20 years, BAM is operating the BAMline at the synchrotron BESSY II in Berlin Adlershof. During this time, the complexity of the setup and the amount of data generated have multiplied. To increase the effectiveness and in preparation for BESSY III, algorithms from the field of machine learning are increasingly used.
In this paper, several examples in the areas of beamline alignment and measurement time optimization based on Bayesian optimization (BO) with Gaussian processes (GP) are presented. BO is a method for finding the global optimum of a function using a probabilistic model represented by a GP. The advantage of this method is that it can handle high-dimensional problems, does not depend on the initial estimate, and also provides uncertainty estimates.
After a short introduction to BO and GP, the first example is the automatic alignment of our double multilayer monochromator (DMM). To achieve optimal performance, up to three linear and two angular motor positions have to be optimized. To achieve this with a grid scan, at least 100^5 measurement points would be required. Assuming that all positions can be aligned independently, 100*5 points are still necessary. We show that with BO and GP less than 100 points are sufficient to achieve equal or better results.
The second example is the optimization of measurement time in XRF scanning. Here we will show the advantage of the BO GP approach over point-by-point scanning. As can be seen in Fig. 1, the number of points required and thus the measurement time can be reduced by a factor of 50, while the loss in image quality is acceptable. The advantages and limitations of this approach will be discussed.