ALBERT

Description: Two-dimensional transition-metal dichalcogenides (TMDs) possess interesting catalytic properties for the electrochemical-assisted hydrogen-evolution reaction (HER). We used niobium diselenide (NbSe2) as a representative TMD, and prepared single-layer NbSe2 porous nanosheets (PNS) by a double-sonication liquid-phase exfoliation, with H2O2 as a pore-forming agent. The single-layer NbSe2 PNS were drop-cast on carbon foam (CF) to fabricate a three-dimensional robust NbSe2 PNS/CF electrode. The NbSe2 PNS/CF electrode exhibits a high current density of −50 mA cm−2 with an overpotential of 148 mV and a Tafel slope of 75.8 eV dec−1 for the HER process. Little deactivation is detected in continuous CV testing up to 20,000 cycles, which suggests that this novel NbSe2 PNS/CF is a promising catalytic electrode in the HER application. The porous structure of single-layer NbSe2 nanosheets can enhance the electrochemical performance compared with that of pore-free NbSe2 nanosheets. These findings illustrate that the single-layer NbSe2 PNS is a potential electrocatalytic material for HER. More importantly, the electrochemical performance of the NbSe2 PNS/CF expands the use of two-dimensional TMDs in electrocatalysis-related fields.

Description: The high-speed-train pantograph is a complex structure that consists of different rod-shaped and rectangular surfaces. Flow phenomena around the pantograph are complicated and can cause a large proportion of aerodynamic noise, which is one of the main aerodynamic noise sources of a high-speed train. Therefore, better understanding of aerodynamic noise characteristics is needed. In this study, the large eddy simulation (LES) coupled with the acoustic finite element method (FEM) is applied to analyze aerodynamic noise characteristics of a high-speed train with a pantograph installed on different configurations of the roof base, i.e. flush and sunken surfaces. Numerical results are presented in terms of acoustic pressure spectra and distributions of aerodynamic noise in near-field and far-field regions under up- and down-pantograph as well as flushed and sunken pantograph base conditions. The results show that the pantograph with the sunken base configuration provides better aerodynamic noise performances when compared to that with the flush base configuration. The noise induced by the down-pantograph is higher than that by the up-pantograph under the same condition under the pantograph shape and opening direction selected in this paper. The results also indicate that, in general, the directivity of the noise induced by the down-pantograph with sunken base configuration is slighter than that with the flush configuration. However, for the up-pantograph, the directivity is close to each other in Y-Z or X-Z plane whether it is under flush or sunken roof base condition. However, the sunken installation is still conducive to the noise environment on both sides of the track.

Description: Currently, haze removal of images captured at night for foggy scenes rely on the traditional, prior-based methods, but these methods are frequently ineffective at dealing with night hazy images. In addition, the light sources at night are complicated and there is a problem of inconsistent brightness. This makes the estimation of the transmission map complicated in the night scene. Based on the above analysis, we propose an autoencoder method to solve the problem of overestimation or underestimation of transmission captured by the traditional, prior-based methods. For nighttime hazy images, we first remove the color effect of the haze image with an edge-preserving maximum reflectance prior (MRP) method. Then, the hazy image without color influence is input into the self-encoder network with skip connections to obtain the transmission map. Moreover, instead of using the local maximum method, we estimate the ambient illumination through a guiding image filtering. In order to highlight the effectiveness of our experiments, a large number of comparison experiments were conducted between our method and the state-of-the-art methods. The results show that our method can effectively suppress the halo effect and reduce the effectiveness of glow. In the experimental part, we calculate that the average Peak Signal to Noise Ratio (PSNR) is 21.0968 and the average Structural Similarity (SSIM) is 0.6802.

Description: The wake region of high-speed trains is an area of complex turbulent flow characterized by the periodic generation and shedding of vortices, which causes discomfort to passengers and affects the stability and safety of the train. In this study, the unsteady characteristics of the wake flows of three 1:1 scale China Railway High-Speed 380A (CRH380A) high-speed train models with different degrees of simplification were numerically investigated using the improved delayed detached eddy simulation (IDDES) method. Analyses of the aerodynamic forces, train-induced slipstream, and turbulent kinetic energy (TKE) were conducted to determine the effects of the bogies on the wake flow of the high-speed train. It was found that the existence of bogies on the bottom of the train, especially the last bogie, not only enhanced the wake flow but also introduced large perturbances into the wake flow. Moreover, the generation and evolution of the vortices in the wake flows were determined by analyzing the instantaneous flow fields and coherent flow structures that were obtained by the dynamic mode decomposition (DMD) method. The results showed that a pair of large, counter-rotating streamwise vortices in the real model of the high-speed train was generated by the cowcatcher and their intensity was significantly enhanced by perturbances that were introduced by the bogies on the bottom of the train.

Description: High dimensional image classification is a fundamental technique for information retrieval from hyperspectral remote sensing data. However, data quality is readily affected by the atmosphere and noise in the imaging process, which makes it difficult to achieve good classification performance. In this paper, multiple kernel learning-based low rank representation at superpixel level (Sp_MKL_LRR) is proposed to improve the classification accuracy for hyperspectral images. Superpixels are generated first from the hyperspectral image to reduce noise effect and form homogeneous regions. An optimal superpixel kernel parameter is then selected by the kernel matrix using a multiple kernel learning framework. Finally, a kernel low rank representation is applied to classify the hyperspectral image. The proposed method offers two advantages. (1) The global correlation constraint is exploited by the low rank representation, while the local neighborhood information is extracted as the superpixel kernel adaptively learns the high-dimensional manifold features of the samples in each class; (2) It can meet the challenges of multiscale feature learning and adaptive parameter determination in the conventional kernel methods. Experimental results on several hyperspectral image datasets demonstrate that the proposed method outperforms several state-of-the-art classifiers tested in terms of overall accuracy, average accuracy, and kappa statistic.

Description: In this paper, a spectral-spatial convolution neural network with Siamese architecture (SSCNN-S) for hyperspectral image (HSI) change detection (CD) is proposed. First, tensors are extracted in two HSIs recorded at different time points separately and tensor pairs are constructed. The tensor pairs are then incorporated into the spectral-spatial network to obtain two spectral-spatial vectors. Thereafter, the Euclidean distances of the two spectral-spatial vectors are calculated to represent the similarity of the tensor pairs. We use a Siamese network based on contrastive loss to train and optimize the network so that the Euclidean distance output by the network describes the similarity of tensor pairs as accurately as possible. Finally, the values obtained by inputting all tensor pairs into the trained model are used to judge whether a pixel belongs to the change area. SSCNN-S aims to transform the problem of HSI CD into a problem of similarity measurement for tensor pairs by introducing the Siamese network. The network used to extract tensor features in SSCNN-S combines spectral and spatial information to reduce the impact of noise on CD. Additionally, a useful four-test scoring method is proposed to improve the experimental efficiency instead of taking the mean value from multiple measurements. Experiments on real data sets have demonstrated the validity of the SSCNN-S method.

Description: (1) Background: The ratio of railway tunnel to line is larger, which produces tunnel entrance and exit effect, aerodynamic resistance, and sudden pressure changes. When the train passes through the tunnels at high-speed, the interaction between the pantograph on it and its surrounding air intensifies and the coupling effects between the pantograph and tunnel become more significant; (2) Methods: A coupling method between aerodynamics and multi-body dynamics is proposed based on hybrid meshing and grid motion. The layered grid motion method is combined with the viscous mesh deformation method with swift, effective data exchange. The significant coupling effects between the pantograph and tunnel are revealed; (3) Results: The influence laws and evolution mechanism of running speed as it affects important service characteristics and behaviors of the pantograph are accurately quantified. Noteworthy factors include the temporal characteristics of panhead aerodynamic lift, the contact force between the pantograph and catenary, vertical displacement and acceleration of the contact strip, the phase diagram of the contact strip, and various frequency-domain characteristics. The action mechanism of running speed on the coupling effect between the pantograph and tunnel is comprehensively and accurately revealed by the proposed method; (4) Conclusions: The larger service characteristics amplitudes of the high-speed pantograph appear at low frequencies and are not multiple frequencies of the basic frequency. By comparisons, the coupling calculation results are closer to the test results than the non-coupling results regardless of the maximum, minimum, or mean.