Search

Background Feature matching technology is vital to establish the association between virtual and real objects in virtual reality and augmented reality systems. Specifically, it provides them with the ability to match a dynamic scene. Many image matching methods, of which most are deep learning-based, have been proposed over the past few decades. However, vessel fracture, stenosis, artifacts, high background noise, and uneven vessel gray-scale make vessel matching in coronary angiography extremely difficult. Traditional matching methods perform poorly in this regard. Methods In this study, a topological distance-constrained feature descriptor learning model is proposed. This model regards the topology of the vasculature as the connection relationship of the centerline. The topological distance combines the geodesic distance between the input patches and constrains the descriptor network by maximizing the feature difference between connected and unconnected patches to obtain more useful potential feature relationships. Results Matching patches of different sequences of angiographic images are generated for the experiments. The matching accuracy and stability of the proposed method is superior to those of the existing models. Conclusions The proposed method solves the problem of matching coronary angiographies by generating a topological distance-constrained feature descriptor.

Because of their superior mechanical performance at ultra-high temperatures, refractory niobium–silicon-based alloys are attractive high-temperature structural alloys, particularly as structural components in gas turbine engines. However, the development of niobium–silicon-based alloys for applications is limited because of the trade-off between room temperature fracture toughness and high-temperature strength. Here, we report on the fabrication of a Nb-18Si alloy with dispersion of hafnium carbide (HfC) particles through selective laser melting (SLM). XRD and SEM-BSE were used to examine the effects of scanning speed on the microstructure and the phase structure of the deposited Nb-18Si-5HfC alloy. The results show that when the scanning speed rises, the solid solubility of the solid solution improves, the interlamellar spacing of eutectics slowly decrease into nano-scale magnitude, and the corresponding hafnium carbide distribution becomes more uniform. We also discover the hafnium carbide particles dispersion in the inter-lamella structure, which contributes to its high fracture toughness property of 20.7 MPa∙m1/2 at room temperature. Hardness and fracture toughness are simultaneously improved because of the control of microstructure morphology and carbide distribution.

Social service ability is now an important indicator to measure the strength and level of running a university. With the diversified development of society, the society's demand for service-oriented talents is gradually expanding. How to optimize the ways of social service training and teaching to cultivate a society for students Service function, improving students' learning enthusiasm and ability to actively participate in social service training is the focus and difficulty that universities need to pay attention to and solve. This article attempts to introduce artificial intelligence technology to this, by using artificial intelligence's human-computer interaction experience and multimedia teaching methods to simulate the actual scene of social service training, while optimizing teachers’ teaching methods, it also improves students’ interest in learning and ability to actively participate. Aiming at achieving a win-win situation between "teaching" and "learning". The experimental survey showed that: nearly 66.67% of the students said they were very satisfied with the improved teaching method, while 0% of the students said they were not satisfied with the improved teaching method; at the same time, the use of optimized teaching methods has improved the evaluation scores of students in all grades. The first year students are the most significant. Their social service curriculum evaluation scores in the next semester have increased from more than 80 points in the previous semester to more than 90 points, an average increase per person scored more than 5 points. This shows that artificial intelligence has a positive effect on optimizing and improving the teaching methods of college social service training.

Nano-sized Y2O3:Yb3+, Er3+ phosphors were synthesized by sol-gel method and characterized by X-ray diffraction, scanning electron microscope, near-infrared and upconversion fluorescence spectroscopy. The phosphors calcined at 1100 °C for 2 h were confirmed to be cubic Y2O3 phase with well-crystallized particles. With the introduction of Ce3+ ions at proper doping concentration, the 1.534 μm emission of Er3+ was found to be greatly enhanced, while the green and red upconversion emission intensity obviously decreased. The enhancement of 1.534 μm emission could be attributed to the phonon-assisted energy transfer process between Ce3+ and Er3+ ions.

Well-crystalline β-NaYF4:Yb3+, Ho3+, Tm3+ nanoparticles were synthesized by sol–gel method using isopropyl alcohol [(CH3)2CHOH] as a complexing agent. The samples were characterized by X-ray diffraction, scanning electron microscopic analysis and fluorescence spectrum analysis methods. Under the excitation of 980 nm laser diode (LD), the samples displayed bright upconversion luminescence (UCL), which was generated from the energy level transition of Ho3+ and Tm3+ ions. With the increase of Tm3+, Ho3+ and Yb3+-doping concentration, the UCL intensity of blue, green and red light emission of the samples varied. Calculation of the CIE color coordinate of the β-NaYF4:Yb3+, Ho3+, Tm3+ nanoparticles revealed that with the adjustment of Tm3+, Ho3+ and Yb3+ doping concentration and the excitation power of 980 nm LD, the multi-color UCL can be realized. Approximately single red light output with the CIE color coordinate of x = 0.545, y = 0.306 and white light output with the CIE color coordinate of x = 0.325, y = 0.320 can be obtained in the synthesized β-NaYF4: Yb3+, Ho3+, Tm3+ nanoparticles.

The hydrodynamics characteristics and heterogeneous structures in an airlift loop reactor(ALR) are analyzed through computational fluid dynamics (CFD) approach. Based on the two-fluid model, a modified cluster-structure-dependent (CSD) drag model with consideration of the effects of cluster is employed and applied to the prediction of the non-uniform flow structure in the airlift loop reactor. In comparison with experimental data and traditional drag model, the present model gives a better prediction. The distributions of local granular temperature in different regions of ALR are shown. The results reveal that the granular temperature in the particle diffluence region is higher than those in the draft tube and annular zones.

On the basis of the Gibbs free energy minimization principle, the dry autothermal reforming performance of crude glycerol in situ hydrogen separation is investigated via thermodynamic analysis. The impact of hydrogen separation fraction on gas composition in product, carbon formation and reaction heat is studied. It can be found that the hydrogen separation promotes the hydrogen production and hinders methane formation. The hydrogen removal is selective to the reduction of carbon deposition, which improves the carbon formation at a low feed CO 2 to glycerol molar ratio and the impact is reverse for high feed CO 2 to glycerol molar ratio. When the reaction temperature varies from 850 K to 900 K, the required oxygen to glycerol molar ratio of thermal neutral condition is obviously increased from 0.15 to 0.4 with hydrogen removal. Meanwhile, the glycerol impurities evaluation indicates that the syngas yield is significantly reduced with the increase of the glycerol impurities. At a high temperature, the hydrogen removal is in favor of the achievement of autothermal process.

The performance of glycerol autothermal reforming in membrane reactors with in situ hydrogen separation is investigated by means of thermodynamic analysis based on Gibbs free energy minimization method. The influence of operational parameters such as reaction temperatures and oxygen to glycerol ratios under different hydrogen separation fractions is evaluated. It is found that reaction temperatures have a more significant impact on hydrogen yield compared to other parameters. The hydrogen removal can promote the hydrogen production obviously and reduce the required temperature of the maximum hydrogen yield. The hydrogen separation effect on carbon formation is discussed. Meanwhile, the performance of ethanol–glycerol mixture autothermal reforming process is also analyzed.

Glycerol reforming technology provides an efficient means for the hydrogen production industry and the utilization of biodiesel-derived crude glycerol. Fluidized bed reactors have potential in interphase mass/heat transfer and mitigating catalyst deactivation. In this study, computational fluid dynamics is used to investigate the performance of glycerol steam reforming in a fluidized bed system. A bubble-structure-dependent drag model is implemented into the two-fluid model to reflect the bubble effect. A three-dimensional simulation is conducted. The model is verified by experimental data. By analysis of the influence of operating conditions and property parameters, it is found that the reaction temperature plays a vital role in the enhancement of hydrogen production.

Zinc complexes L2Zn and LZnX were prepared by template synthesis from ZnX2, N1,N1-dimethylbenzene-1,2-diamine and substituted salicylaldehyde (SA). A systematic exploration of this synthetic route revealed the subtle role of choice of Zn source and SA substitution pattern on the reaction stoichiometry and enabled synthesis of pure Zn complexes. Use of poorly coordinating counterions (such as CF3SO3−) and apolar substituents on the SA component favours formation of pentacoordinate L2Zn, while more coordinating counterions tend to produce LZnX, (LZnX)2 or even binuclear LZn(O2CR)Zn(O2CR)2 not reported before. The complexes (LZnX)2 and LZn(O2CR)·Zn(O2CR)2 (L = L2 or L4) could be activated with BnOH/LiMe for ring-opening polymerization of e-caprolactone; bis(ligand) complexes L2Zn show low or vanishing activity when activated either by BnOH or BnOH/LiMe.

The steam absorption process in a fluidized bed of binary mixture is investigated to study the mass transfer performance of active particles surrounding inert particles by means of CFD approaches. A bubble-based drag model is incorporated to the multi-fluid model to account for the bubble effect on mixing and segregation behaviors of binary mixture in a fluidized bed system. The predictions using different drag models are compared with experimental results. The effect of mixing and segregation behaviors of inert particles and active particles on mass transfer performance is evaluated. The sensitivity of mass transfer process to operating conditions is also analyzed. The results reveal that the bubble-based drag model can give a better prediction with measured data. The enhancement of mixing degree of inert particles and active particles can promote the mass transfer performance.

Polystyrene (PS) nanoparticles coated by BSA, hereafter denoted as PS/BSA, were prepared and chemically immobilized for the first time onto a capillary inner wall for open-tubular CEC (OTCEC). EOF and scanning electron micrography were used to characterize the prepared nanoparticle-coated capillaries. To investigate the performance of the prepared columns in OTCEC, chiral separation of d,l-tryptophan (dl-Trp) was performed in monolayer BSA-modified capillary and PS/BSA nanoparticle-coated columns. The results indicated that the nanoparticle-modified column afforded a higher resolution compared with the monolayer type. Rapid enantioseparation of dl-Trp (within 3 min) was achieved with the PS/BSA-immobilized column using an electroosmotic pump-assisted CEC. Enantiomer separations of other compounds like dl-tyrosine and warfarin were also achieved with the column. Besides, run-to-run and column-to-column repeatabilities of the PS/BSA-coated column in the chiral separation were systematically introduced.