Filtern
Dokumenttyp
- Zeitschriftenartikel (16) (entfernen)
Schlagworte
- Machine learning (4)
- Clustering (2)
- Concrete evaluation (2)
- Corrosion (2)
- Data fusion (2)
- Honeycombing (2)
- Materials informatics (2)
- Sequential learning (2)
- Alkali Activated Concrete (1)
- Alkali-activated binders (1)
Organisationseinheit der BAM
- 8 Zerstörungsfreie Prüfung (12)
- 8.0 Abteilungsleitung und andere (12)
- 7 Bauwerkssicherheit (3)
- 5 Werkstofftechnik (2)
- 5.2 Metallische Hochtemperaturwerkstoffe (2)
- 7.6 Korrosion und Korrosionsschutz (2)
- 1 Analytische Chemie; Referenzmaterialien (1)
- 1.3 Instrumentelle Analytik (1)
- 5.1 Mikrostruktur Design und Degradation (1)
- 7.7 Modellierung und Simulation (1)
Paper des Monats
- ja (1)
An Ontology-Based Approach to Enable Data-Driven Research in the Field of NDT in Civil Engineering
(2021)
Although measurement data from the civil engineering sector are an important basis for scientific analyses in the field of non-destructive testing (NDT), there is still no uniform representation of these data. An analysis of data sets across different test objects or test types is therefore associated with a high manual effort. Ontologies and the semantic web are technologies already used in numerous intelligent systems such as material cyberinfrastructures or research databases. This contribution demonstrates the application of these technologies to the case of the 1H nuclear magnetic resonance relaxometry, which is commonly used to characterize water content and porosity distri-bution in solids. The methodology implemented for this purpose was developed specifically to be applied to materials science (MS) tests. The aim of this paper is to analyze such a methodology from the perspective of data interoperability using ontologies. Three benefits are expected from this ap-proach to the study of the implementation of interoperability in the NDT domain: First, expanding knowledge of how the intrinsic characteristics of the NDT domain determine the application of semantic technologies. Second, to determine which aspects of such an implementation can be improved and in what ways. Finally, the baselines of future research in the field of data integration for NDT are drawn.
This paper studies the dispersion of GPR wave’s phase velocity at different wideband frequencies in plywood and concrete with varying moisture content. This study makes use of two GPR antennas with 2 GHz centre frequency operating in wide angle reflection and refraction (WARR) mode and with computation of spectral analysis of the surface wave (SASW). Computation of phase velocities is based on the Acquisition of the cross-power spectrum and phase unwrap of two distorted ground waves at positions Closer to and farther away from the transmitting antenna. The velocities of the ground waves are found to experience greater dispersion in low frequency regimes within the effective frequency bandwidths determined and thresholded by time–frequency analysis (TFA) and coherence plotting of the ground waves.
This study validates not only the methodology, but also identifies the optimal distance between the first (Rx1) and second (Rx2) receivers as k/2, which is based on a fixed transmitter (Tx) minus the first Receiver (Rx1) distance. It serves as an indication of changeable separation distance when other lower frequency GPR is used because the distances of Tx-Rx1 and Rx1-Rx2 are wavelength dependent and thus also frequency dependent. Effects of moisture contents and chloride contamination in concrete were also characterized according to the dispersion plots . that wave traelling in lower frequencies in GPR wave is much decelerated than that in higher frequencies. This research also contributes to the building of the ‘‘GPRWARR machine” suggested in Annan and Jackson (2017) [1], within which the effects of wave Dispersion on phase velocity can be inversely modelled to characterize variations in the material properties of infrastructure as a means of detecting surface damage.
Alkali-activated binders (AAB) can provide a clean alternative to conventional cement in terms of CO2 emissions. However, as yet there are no sufficiently accurate material models to effectively predict the AAB properties, thus making optimal mix design highly costly and reducing the attractiveness of such binders. This work adopts sequential learning (SL) in high-dimensional material spaces (consisting of composition and processing data) to find AABs that exhibit desired properties. The SL approach combines machine learning models and feedback from real experiments. For this purpose, 131 data points were collected from different publications. The data sources are described in detail, and the differences between the binders are discussed. The sought-after target property is the compressive strength of the binders after 28 days. The success is benchmarked in terms of the number of experiments required to find materials with the desired strength. The influence of some constraints was systematically analyzed, e.g., the possibility to parallelize the experiments, the influence of the chosen algorithm and the size of the training data set. The results show the advantage of SL, i.e., the amount of data required can potentially be reduced by at least one order of magnitude compared to traditional machine learning models, while at the same time exploiting highly complex information. This brings applications in laboratory practice within reach.
For the closure of radioactive waste disposal facilities engineered barriers- so called “drift seals” are used. The purpose of these barriers is to constrain the possible infiltration of brine and to prevent the migration of radionuclides into the biosphere. In a rock salt mine a large scale in-situ experiment of a sealing construction made of salt concrete was set up to prove the technical feasibility and operability of such barriers. In order to investigate the integrity of this structure, non-destructive ultrasonic measurements were carried out.
Therefore two different methods were applied at the front side of the test-barrier:
1 Reflection measurements from boreholes
2 Ultrasonic imaging by means of scanning ultrasonic echo methods This extended abstract is a short version of an article to be published in a special edition of ASCE Journal that will briefly describe the sealing construction, the application of the non-destructive ultrasonic measurement methods and their adaptation to the onsite conditions -as well as parts of the obtained results. From this a concept for the systematic investigation of possible contribution of ultrasonic methods for quality assurance of sealing structures may be deduced.
We present the results of several machine learning (ML)- inspired data fusion algorithms applied to multi-sensory nondestructive testing (NDT) data. Our dataset consists of Impact-Echo (IE), Ultrasonic Pulse Echo (US) and Ground Penetrating Radar (GPR) data collected on large-scale concrete specimens with built–in simulated honeycombing defects. The main objective is to improve the detectability of honeycombs by fusing the information from the three different sensors. We describe normalization, feature detection and optimal feature selection. We have used unsupervised and supervised ML, i.e., classification and clustering, for data fusion. We demonstrate the advantage of data fusion in reducing the false positives up to 10% compared to the best single sensor, thus, improving the detectability of the defects. The methods were evaluated on a concrete specimen. The effectiveness of the proposed approach was demonstrated on a separate full-scale concrete specimen. The results indicate the transportability of the conclusions from one specimen to the other.
Potentialfeldmessung (PM) ist die beliebteste Methode der Zerstörungsfreien Prüfung (ZfP) zur Lokalisierung von aktiver Betonstahlkorrosion. PM wird durch Parameter wie z. B. Feuchtigkeits- und Chloridgradienten im Bauteil beeinflusst, so dass die Sensitivität gegenüber der räumlich sehr begrenzten, aber gefährlichen Lochkorrosion gering ist. Wir zeigen in dieser Studie, wie zusätzliche Messinformationen mit Multisensor-Datenfusion genutzt werden können, um die Detektionsleistung zu verbessern und die Auswertung zu automatisieren. Die Fusion basiert auf überwachtem maschinellen Lernen (ÜML). ÜML sind Methoden, die Zusammenhänge in (Sensor-) Daten anhand vorgegebener Kennzeichnungen (Label) erkennen. Wir verwenden ÜML um „defekt“ und „intakt“ gelabelte Bereiche in einem Multisensordatensatz zu unterscheiden. Unser Datensatz besteht aus 18 Messkampagnen und enthält jeweils PM-, Bodenradar-, Mikrowellen-Feuchte- und Wenner-Widerstandsdaten. Exakte Label für veränderliche Umweltbedingungen wurden in einer Versuchsanordnung bestimmt, bei der eine Stahlbetonplatte im Labor kontrolliert und beschleunigt verwittert. Der Verwitterungsfortschritt wurde kontinuierlich überwacht und die Korrosion gezielt erzeugt. Die Detektionsergebnisse werden quantifiziert und statistisch ausgewertet. Die Datenfusion zeigt gegenüber dem besten Einzelverfahren (PM) eine deutliche Verbesserung. Wir beschreiben die Herausforderungen datengesteuerter Ansätze in der zerstörungsfreien Prüfung und zeigen mögliche Lösungsansätze.
This work presents machine learning-inspired data fusion approaches to improve the non-destructive testing of reinforced concrete. The principal effects that are used for data fusion are shown theoretically. Their effectiveness is tested in case studies carried out on largescale concrete specimens with built-in chloride-induced rebar corrosion. The dataset consists of half-cell potential mapping, Wenner resistivity, microwave moisture and ground penetrating radar measurements. Data fusion is based on the logistic Regression algorithm.
It learns an optimal linear decision boundary from multivariate labeled training data, to separate intact and defect areas. The training data are generated in an experiment that simulates the entire life cycle of chloride-exposed concrete building parts. The unique possibility to monitor the deterioration, and targeted corrosion initiation, allows data labeling.
The results exhibit an improved sensitivity of the data fusion with logistic regression compared to the best individual method half-cell potential.
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.