Search

Multispectral Sentinel-2 (S-2) images are a valuable source of Earth observation data; however, spatial resolution of their spectral bands limited to 10-, 20-, and 60-m ground sampling distance (GSD) remains insufficient in many cases. This problem can be addressed with super-resolution (SR), aimed at reconstructing a high-resolution (HR) image from a low-resolution (LR) observation. For S-2, spectral information fusion allows for enhancing the 20- and 60-m bands to the 10-m resolution. Also, there were attempts to combine multitemporal stacks of individual S-2 bands; however, these two approaches have not been combined so far. In this article, we introduce DeepSent—a new deep network for super-resolving multitemporal series of multispectral S-2 images. It is underpinned with information fusion performed simultaneously in the spectral and temporal dimensions to generate an enlarged multispectral image (MSI). In our extensive experimental study, we demonstrate that our solution outperforms other state-of-the-art techniques that realize either multitemporal or multispectral data fusion. Furthermore, we show that the advantage of DeepSent results from how these two fusion types are combined in a single architecture, which is superior to performing such fusion in a sequential manner. Importantly, we have applied our method to super-resolve real-world S-2 images, enhancing the spatial resolution of all the spectral bands to 3.3-m nominal GSD, and we compare the outcome with very HR WorldView-2 images. We have made our implementation publicly available, and we expect it will increase the possibilities of exploiting super-resolved S-2 images in real-life applications.

By use of physical simulations, it was studied the influence of finish rolling temperature (from 820 °C to 970 °C) on the microstructural and mechanical properties of seamless tubes with a different wall thickness (from 6.3 to 40 mm) – in the state after rolling as well as after quenching and tempering. In laboratory conditions, by use of the Simulator HDS-20, the bloom piercing and rolling of the seamless tubes from 25CrMo4 low-alloy steel in a pilger mill were in a simplified way simulated. The wall thickness of the tube influenced the total deformation of specimens at anisothermal multi-pass plain-strain compression tests as well as the final cooling rate. The quenching and tempering of the deformed specimens was subsequently performed with use of the electric resistance furnaces. The finish rolling temperature had only insignificant effect on the resulting properties. Markedly lower hardness was obtained only after the simulation of tube production with the wall thickness of 40 mm contrary to the wall thickness of 6.3 and 20 mm. Structural variations of the specimens after rolling simulations were more or less overlapped by the subsequent quenching from the temperature of 850 °C and tempering at the temperature of 680 °C.

Phase-pure crystalline Bi2Se3and Bi2Te3nanoparticles are formed in reactions of [C4C1Im]3[Bi3I12] (C4C1Im = 1-butyl-3-methylimidazolium) with [C4C1Pyr][ESiMe3] (E = Se or Te; C4C1Pyr = 1-butyl-1-methylpyrrolidinium) in the ionic liquid (IL) [C4C1Im]I. The resulting crystalline tetradymite-type nanoparticles exhibit stoichiometric Bi:E (E = Se or Te) molar ratios (2:3). Because all synthetic steps were performed under strict inert gas conditions, the surfaces of the Bi2Se3and Bi2Te3nanoparticles are free of metal oxide species. As proven by infrared and X-ray photoelectron spectroscopy analyses, the nanoparticle surfaces reveal only minor organic contamination from solvent residues ([C4C1Im]I). The nanomaterials show high Seebeck coefficients of −124 μV K–1(Bi2Se3) and −155 μV K–1(Bi2Te3) and feature high electrical conductivities (328 and 946 S cm–1, respectively) at the highest tested temperature (240 °C). The corresponding thermal conductivities (0.8 and 2.3 W m–1K–1, respectively, at 30 °C) are comparable to those of single crystals and recently reported ab initiocalculations, which is in remarkable contrast to typical findings of nanograined bulk materials obtained from compacted nanoparticles. These findings emphasize the low level of impurities, surface contamination, and, in general, defects produced by the synthetic approach reported here. The figure of merit in the in-plane direction of the compacted pellets reached peak values 0.45 for Bi2Se3and 0.4 for Bi2Te3.

Using the laboratory rolling mill with smooth rolls, piercing, as well as rolling in a pilger mill of the seamless tubes with diameter 273 mm from the HSLA steel microalloyed with vanadium steel was simulated. Influence of the wall thickness (6.3 – 40 mm) and finish rolling temperature on the final structural and mechanical properties was investigated. Necessary temperatures of the phase transformations in the course of cooling were determined by dilatometric tests. Based on the dilatometry results, finish rolling temperatures were reduced. Lower rolling temperatures yielded in a relative grain refinement. Effect of the finish rolling temperature did not have any marked impact on the tensile tests results. Strength properties decreased only slightly with the increasing wall thickness and the plastic properties were not influenced significantly by this parameter. The positive effect of the reduced finishing temperature appeared markedly in the results of impact tests performed at room temperature only. Notch toughness was increased by approx. 25 % in the case of the wall thickness of not less than 20 mm.

The CCT and DCCT diagrams of steel C60 (with approx. 0.6 % C) were constructed on the basis of dilatation tests with and/or without an influence of the previous deformation and they were then compared, order to make an evaluation of the influence of the previous deformation on the phase transformation kinetics. For the execution of the experiment, the dilatation module of the plastometer Gleeble 3800 was used. The accuracy of the diagrams was faced with metallographic analyses and measurements. The previous deformation expressly retarded a bainite transformation and slightly accelerated ferrite and pearlite transformations. The martensite start temperature was practically not influenced by the previous deformation; however, the applied deformation caused the creation of the martensite at lower cooling rates.

Susceptibility to cracking of the as-cast C-Mn-Cr-Nb steel was studied by laboratory rolling. The variable parameters were the heating temperature (1150 - 1340 °C) as well as the rolling temperature (950 - 1150 °C). Final microstructure of the free-cooled samples was constituted by bainite, pearlite and ferrite with different morphology and various contribution. Deformation temperature below 1000 °C yielded in the incomplete recrystallization of austenite. Surface cracks originated preferentially on the austenite grains boundaries. Size of the present particles (inclusions and precipitates) varied from 101 nm to 101 μm. SEM and EDS analysis revealed that the inclusions ware mostly of the MnS type. TEM analysis confirmed that the grain boundaries were not enriched by any particles. In addition to the Fe3C particles, the discoid niobium carbide particles with approximately 40 nm diameter and 10 nm thickness were detected. These small particles were not connected by any notable pinning of dislocations.

The samples intended for the study of static recrystallization on plastometer Gleeble were prepared from the laboratory castings of iron aluminide containing 24.6 Al 0.17 Mn 0.16 Zr 0.026 B 0.004 C (in wt. %, remainder Fe). Nevertheless, the structure analysis discovered that the results were excessively influenced by the huge heterogeneity of the as-cast microstructure, mostly of the grain size. Combination of the hot forming and recrystallization process during the long-term high-temperature annealing was selected for the necessary structure homogenization. As the tested intermetallic alloy is extremely brittle and susceptible to surface cracking, the original method was applied for its processing. The method consists in hot rolling in the protective capsules welded from the ferritic stainless steel sheet. The castings were rolled to 2/3 of their thickness by 4 reductions with the inter-stage heating, and then annealed at the temperature of 1200 °C in the vacuum furnace for several periods. Metallographic analysis revealed that annealing lasting 7 hours was essential for the uniform coarsening of the recrystallized grains. Material processed in this way proved successful for the subsequent metallographical study of static recrystallization.

The aim of works was to describe mathematically the kinetics of static recrystallization of the alloy type Fe–40at.%Al–Zr–B (with 24.6 Al–0.01 B–0.18 Zr–0.01 C in wt.%) with the coarse-grained structure. The microstructure of the laboratory castings made of this extremely brittle alloy was homogenized by hot rolling of the material in the protective capsules and by the long-term annealing at 1200°C. An initial microstructure with average grain size 0.77±0.27mm was obtained. Based on the isothermal plastic deformation tests and EBSD analysis, the static recrystallization kinetics of the prepared coarse-grained B2 iron aluminide after strain 0.20 was mathematically described. Recrystallized fraction depends on deformation/annealing temperature (900–1100°C) as well as on annealing time. The activation energy of static recrystallization was calculated as 255kJmol−1. Competition between dynamic recovery and static recrystallization was proved after strain 0.35 and annealing temperature 1100°C. Static recrystallization starts relatively easily in the studied alloy, but a very long-term annealing is quite necessary for the complete course of recrystallization. The mean size of recrystallized grains falls with the decreasing annealing temperature (0.47±0.15mm for temperature 1100°C, and 0.22±0.04mm for 900°C). Even at a temperature of 1200°C the annealing after deformation should last approx. 1min for obtaining the fully recrystallized microstructure. That is why the standard hot forming technologies should be combined by an interpass annealing in order to refine sufficiently the coarse grains.

The objective of investigation was to determine the influence of chosen cooling rates after finish rolling on final microstructural and mechanical properties of the laboratory rolled products from steel 42CrMo4. Metallographic analysis showed that microstructure of rolled products, which were after finish rolling cooled down in accelerated manner by water sprays, was composed mostly by hardening phases and by smaller amount of ferrite. Microstructure of the rolled product cooled down in furnace in decelerated manner was formed by pearlitic blocks with minority occurrence of ferrite. Laboratory rolled products cooled down by more complicated modes, which consisted of combination of their cooling by water sprays with subsequent annealing in furnace, showed different phase morphology and distinct band structure. Results of tensile test have proved that the applied modes of cooling of rolled products by accelerated manner had only very small influence on their final mechanical properties. Considerably cooling the rolled product by decelerated manner in furnace resulted in drop yield strength, but on the other hand in an increase of ductility. More complicated modes of cooling the rolled products, which comprised also their annealing, showed the possibilities of significant influencing the strength, and particularly the plastic properties of the steel 42CrMo4.

On the basis of the uniaxial compression tests realized in the temperature range from 800 °C to 1200 °C and at strain rates from 0.05 s-1 to 30 s-1 the value of the activation energy at hot forming of coarse-grained as-cast iron aluminide Fe40at.%Al-Zr-B was calculated as Q = 502 kJ·mol-1. With use of this material constant, relatively precise mathematical description of the maximum flow stress and the corresponding strain in dependence on the Zener-Hollomon parameter was possible. As was discovered by metallographical analysis, at high values of the Zener-Hollomon parameter slight decrease in flow stress occurs even without significant course of dynamic recrystallization.

AbstractInfluence of the methodology of mathematical processing of experimental data on the calculated value of the activation energy in hot forming was evaluated for three very different alloys: iron aluminide, low-alloyed steel and magnesium alloy. Application of the mean strain rates is appropriate but calculations with use of nominal strain rates of the plastometric tests do not give significantly different results. Calculations based on the processing of the simplified power dependence between strain rate and peak stress can give the results differing very substantially from those that were gained by the solution of the complex hyperbolic relationship. This has negative impacts, particularly at prediction of the maximum flow stress in the broad range of forming conditions, represented by the Zener-Hollomon parameter.

A physical simulation of the thermomechanical processing of the Mn-Ni-Cr-Mo low-alloy steel was performed in the laboratory rolling mill Tandem in the Institute of Modelling and Control of Forming Processes at VŠB – Technical University of Ostrava. The task was to determine the influence of the finish rolling temperature on the structural and mechanical properties of the rolled products. After different modes of rolling and slow cooling in the furnace, the final structure of the tested samples was in all cases composed of ferrite, bainite and islands of martensite. The finish rolling temperature markedly influenced a part by volume of the individual phases as well as the structure homogeneity. The results of the tensile tests at room temperature indicated that the studied steel did not show any pronounced dependence of the yield stress on the finish rolling temperature in the investigated range of values (750 – 1000 °C). On the other hand, the closely corresponding dependences of the ultimate tensile stress and elongation exhibited a considerable and very complex course, which can be explained mainly by the martensite fraction originating during the last stage of the final air cooling from temperature 600 °C.

In recent years, the study of the hot deformation behavior of various materials is significantly marked by an increasing utilization of artificial neural networks, which are frequently employed for a hot flow curve description. This specific kind of description is commonly solved via a Feed-Forward Multi-Layer Perceptron architecture and rarely via a Radial Basis architecture. Both network architectures are compared to assess their suitability in the process of a hot flow curve description under a wide range of thermomechanical conditions. The performed survey is also aimed on the eventual utilization of corresponding modifications of both studied networks, namely on a Cascade-Forward Multi-Layer Perceptron and Generalized Regression network. The main results have shown that the Feed-Forward Multi-Layer Perceptron architecture represents a good choice if very high accuracy is a crucial goal. However, in the case of this architecture, finding the proper parameters can be time-consuming and the hardware burdensome. On the contrary, for the flow curve description the almost unused Radial Basis network offers a very easy training procedure and significantly shorter computing time under acceptable accuracy. The results of the submitted research should then serve as a background for the selection and following application of a suitable network architecture in the process of solving future flow curve description tasks.

AbstractBased on the measurement of roll forces during the laboratory hot rolling of flat samples graded in thickness, the new phenomenological type of mean flow stress model was developed and applied on plain-carbon and HSLA steels. The obtained models describe with a very good accuracy the hot deformation resistance characteristics in the temperature range 1123 to 1463 K, large effective strain, and strain rate in the useful range of approximately 10 to 100 s−1. Difficulty in the mathematical description of the influence of temperature on mean flow stress in the wide range of temperature by a single equation was solved by introducing another constant in the temperature member of the conventional model. The newly proposed model solves by phenomenological means a frequent problem of heteroscedasticity of relative deviations between the calculated and experimental values of mean flow stress values depending on temperature. It becomes more reliable from the viewpoint of the operational application, e.g. fast prediction of mean flow stress values and power/force parameters necessary in the steering systems of hot rolling mills.

The paper deals with rolling conditions, microstructure, as well as basic mechanical properties of four different high manganese steels after laboratory melting, casting, and hot and cold rolling. The stacking fault energy of heats was in the interval of 85–114mJm-2thanks to the high aluminium contents. The heats with the lowest sum of (C+Mn) supported the highest ferrite volume fraction (up to 45%), while the highest sum of (C+Mn) led to 10% ferrite formation. With lower Al/(C+Mn) ratio and lower ferrite fraction higher rolling forces were necessary to be used. Due to heterogeneities of matrixes and observed aluminium oxide complexes differences in mechanical properties were detected. The best results showed the heat with 0.65% C-29.5% Mn-9.0% Al and a stacking fault energy 114mJm-2.

The focus of this paper is to explore the possibilities of optimizing 3D printed elements produced by FFF/FDM technology based on bending tests according to ČSN EN ISO 178 (64 0607) with variations in the settings of the printing itself. The principle of FFF/FDM is the printing of a continuous fiber made of thermoplastic material, which is applied by machine to the previously printed fiber. There are many combinations of possible print settings, and one of them is the geometry of the inner density with variable density. Their resulting maximum values of the achieved load were then compared with the weight (amount of material used) and printing time. The result is a comparison to achieve economical printing with the greatest possible load capacity.

The processing maps (i.e. power dissipation maps superimposed over instability maps) can be used as avery convenient tool in case of an optimizing of hot forming processes. In this research, processing maps of C45 medium-carbon steel were assembled on the basis of an experimental flow stress dataset. This dataset was acquired via series of uniaxial hot compression tests in the temperature range of 1173 K - 1553 K and the strain rate range of 0.1 s[?]1 - 100 s[?]1. In addition, a predicted flow stress dataset was created with use of an artificial neural network approach - it allowed extending of the experimental dataset with additional temperature levels. The experimentally compiled processing maps have been subsequently enhanced by this additional dataset to encourage the overall information capability. The results have showed that the predicted dataset was useful to reveal additional instability regions in the experimentally assembled processing maps.