Wissenschaftliche Artikel der BAM
Filtern
Erscheinungsjahr
Dokumenttyp
Sprache
- Englisch (2254) (entfernen)
Schlagworte
- Additive manufacturing (69)
- Fluorescence (55)
- Nanoparticles (48)
- SAXS (46)
- Concrete (45)
- Corrosion (37)
- MALDI-TOF MS (36)
- Quantum yield (36)
- Ultrasound (36)
- Non-destructive testing (31)
Organisationseinheit der BAM
- 6 Materialchemie (432)
- 8 Zerstörungsfreie Prüfung (369)
- 1 Analytische Chemie; Referenzmaterialien (350)
- 9 Komponentensicherheit (212)
- 5 Werkstofftechnik (206)
- 7 Bauwerkssicherheit (206)
- 4 Material und Umwelt (184)
- 6.3 Strukturanalytik (164)
- 8.5 Röntgenbildgebung (127)
- 6.1 Oberflächen- und Dünnschichtanalyse (100)
Paper des Monats
- ja (67)
We present a workflow for obtaining fully trained artificial neural networks that can perform automatic particle segmentations of agglomerated, non-spherical nanoparticles from scanning electron microscopy images “from scratch”, without the need for large training data sets of manually annotated images. The whole process only requires about 15 minutes of hands-on time by a user and can typically be finished within less than 12 hours when training on a single graphics card (GPU). After training, SEM image analysis can be carried out by the artificial neural network within seconds. This is achieved by using unsupervised learning for most of the training dataset generation, making heavy use of generative adversarial networks and especially unpaired image-to-image translation via cycle-consistent adversarial networks. We compare the segmentation masks obtained with our suggested workflow qualitatively and quantitatively to state-of-the-art methods using various metrics. Finally, we used the segmentation masks for automatically extracting particle size distributions from the SEM images of TiO2 particles, which were in excellent agreement with particle size distributions obtained manually but could be obtained in a fraction of the time.
Grouting is a universal repair and strengthening technique, which is constantly used for structural remediation of concrete components, trenches, mine subsidence, dam joints, restoration of masonry structures, and geological stabilizations. Having an extremely small particle size of only few microns, ultrafine cements are ideal for grouting applications due to their superior permeability and compressive strength properties of the hardened cement paste compared to that of the less-expensive, but coarser ordinary Portland cements. Supplementary cementitious materials (SCMs) are often used to replace ultrafine cement in order to modify certain properties and to reduce costs. The aim of this experimental study is to investigate the effect of three supplementary materials: microsilica (MS), fly ash (FA), and metakaolin (MK) on the workability, and mechanical properties of an ultrafine cement based grout with a constant water-binder ratio and constant superplasticizer content. Maximum percentages of replacement with ultrafine cement were 6% by volume of cement for MS and 16% for FA, and MK. In general, results suggest that the workability is improved by addition of FA, whereas is reduced, when modified with MS and MK. The compressive strength of grout after cement replacement remains comparable to that of pure cement grout. However, there is a tendency of the MS to positively affect the compressive strength opposite to FA, whereas flexural strength is positively affected by FA. Based on the results, it is evident that grouts with Hägerman cone flow more than 500 mm and compressive strength of more than 90 MPa after 28 days can be produced.
Wire arc additive manufacturing enables the production of near-net shape large-volume metallic components leveraging an established industrial base of welding and cladding technology and adapting it for layer-wise material deposition. However, the complex relationship between the process parameters and resulting mechanical properties of the components still remains challenging. In case of high-strength Al-Mg-Si aluminum alloys, no commercial filler wires are yet available due the high susceptibility of solidification cracking as well as the necessary efforts to obtain acceptable mechanical properties. To address this need, we evaluated a novel filler wire based on AlMg0.7Si doped with a Ti5B1 master alloy to foster fine equiaxed grains within the deposited metal. The correlation between the process parameters and component quality was examined by analyzing the size and distribution of pores as well as the grain morphology. Furthermore, we evaluated the influence of different post-weld heat treatment strategies to achieve mechanical properties corresponding to the reference wrought material. We demonstrated that fine equiaxed grains in the weld metal reduced the susceptibility of solidification cracking significantly. The novel AlMg0.7Si-TiB (S Al 6063-TiB) filler wire facilitated wire arc additive manufacturing of high-strength aluminum components with mechanical properties that were almost as superior as the corresponding wrought base material.
Modern wind turbine rotor blades consist of sandwich shell segments made from glass fiber reinforced polymers. During manufacturing, defects can arise which could lead to failure of the whole component under dynamic mechanical and thermal loads. Hence during operation defects can arise which, if detected, can be repaired locally and in-situ by applying repair patches instead of taking the whole rotor blade down and repair it remotely. This method is much more time and cost effective, since the shut-down time of the energy converter is limited to a minimum. These repair patches can, however, also lead to new defects if not applied optimally. Therefore, it is necessary to control the quality of the repair patches to ensure the best possible restoration of structural integrity of the component. As a rotor blade is an object with a large aspect ratio, X-ray laminography is predestined to provide 3D information of the objective volume. To enhance the amount of information gained from laminographic reconstruction, we use in this study a photon counting and energy discriminating X-ray detector and apply a material decomposition algorithm to the data. By inherently separating the incident spectra within the detection process into two distinct energy bins, the basis material decomposition can provide material resolved images. Choosing glass and epoxy resin as basis materials and numerically solving the inverse dual-energy equation system, the reconstructed laminographic datasets contain highly valuable information about the distribution of the basis materials within the structure. Furthermore, cross- artifacts arising from the limited angle of the projection data can be reduced by this method which allows to investigate structures that were hidden underneath the artefacts.
In this study, a new reliable, economic, and environmentally-friendly one-step synthesis is established to obtain carbon nanodots (CNDs) with well-defined and reproducible photoluminescence (PL) properties via the microwave-assisted hydrothermal treatment of starch and Tris-acetate-EDTA (TAE) buffer as carbon sources. Three kinds of CNDs are prepared using different sets of above mentioned starting materials. The as-synthesized CNDs: C-CND (starch only), N-CND 1 (starch in TAE) and N-CND 2 (TAE only) exhibit highly homogenous PL and are ready to use without need for further purification.
The CNDs are stable over a long period of time (>1 year) either in solution or as freeze-dried powder. Depending on starting material, CNDs with PL quantum yield (PLQY) ranging from less than 1% up to 28% are obtained. The influence of the precursor concentration, reaction time and type of additives on the optical properties (UV-Vis absorption, PL emission spectrum and PLQY) is carefully investigated, providing insight into the chemical processes that occur during CND formation. Remarkably, upon freeze-drying the initially brown CND-solution turns into a non-fluorescent white/slightly Brown powder which recovers PL in aqueous solution and can potentially be applied as fluorescent marker in bio-imaging, as a reduction agent or as a photocatalyst.
The definition of the term “conversion” is discussed for a variety of polymer syntheses. It is demonstrated that in contrast to organic and inorganic chemistry several different definitions are needed in polymer science. The influence of increasing conversion on structure and topology of homo- and Copolymers is illustrated. Chain-growth polymerizations, such as radical polymerization or living anionic polymerizations of vinyl monomers, condensative chain polymerization, two and three-dimensional step-growth polymerizations, ring–ring or chain–chain equilibration and chemical modification of polymers are considered.
High-power laser beam welding became new stimuli within the last 10 years due to the availability of a new generation of high brightness multi kilowatt solid state lasers. In the welding research new approaches have been developed to establish reliable and praxis oriented welding processes meeting the demands of modern industrial applications during this time. The paper focuses on some of the current scientific and technological aspects in this research field like hybrid laser arc welding, simulation techniques, utilization of electromagnetic fields or reduced pressure environment for laser beam welding processes, which contributed to the further development of this technology or will play a crucial role in its further industrial implementation.
The geometry of the melt pool in laser beam welding plays a major role to understand the dynamics of the melt and its solidification behavior. In this study, a butt configuration of 15 mm thick structural steel and transparent quartz glass was used to observe the weld pool geometry by means of high-speed camera and an infrared camera recording. The observations show that the dimensions of the weld pool vary depending on the depth. The areas close to the weld pool surface take a teardrop-shape. A bulge-region and its temporal evolution were observed approximately in the middle of the depth of the weld pool. Additionally, a 3D transient thermal-fluid numerical simulation was performed to obtain the weld pool shape and to understand the formation mechanism of the observed bulging effect. The model takes into account the local temperature field, the effects of phase transition, thermo-capillary convection, natural convection and temperature-dependent material properties up to evaporation temperature. The numerical results showed good accordance and were furthermore used to improve the understanding of the experimentally observed bulging effect.
Weather-dependent passive thermography and thermal simulation of in-service wind turbine blades
(2023)
. To cope with the increase in the manufacturing and operation of wind turbines, wind farm operators need inspection tools that are able to provide reliable information while keeping the downtime low. Current inspection techniques require to stop the wind turbine. This work presents the current progress in the project EvalTherm, in which passive thermography is evaluated as a possible non-destructive inspection tool for operational wind turbine blades (WTBs). A methodology to obtain thermal images of rotating WTBs has been established in this project. However, the quality of the results is heavily dependent on various aspects such as weather conditions, information on the inspected WTB, damage history, etc. In this work, a section of a used WTB is simulated using finite-element modelling (FEM) as well as experimentally tested for evaluating the accuracy of the model. Such a model will provide insight into the potential thermal response of a certain structure (with specific material properties) in given weather (boundary) conditions. The model is able to provide satisfactory predictions of the thermal response of the structure, as well as indicate what thermal contrast(s) result from artificial defects introduced in the structure.
Wavelength-scanning distributed acoustic sensing for structural monitoring and seismic applications
(2019)
We introduce wavelength-scanning coherent optical time domain reflectometry (WS-COTDR) for dynamic vibration sensing along optical fibers. The method is based on spectral shift computation from Rayleigh backscatter spectra. Artificial neural networks (ANNs) are used for fast and high-resolution strain computation from raw measurement data. The applicability of the method is demonstrated for vibration monitoring of a reinforced concrete bridge. We demonstrate another application example for quasi-static and dynamic measurement of ground deformation and surface wave propagation along a dark fiber in a telecommunication cable.