Deep learning technology is rapidly spreading in recent years and has been extensive attempts in the field of Brain-Computer Interface (BCI). Though the accuracy of Motor Imagery (MI) BCI systems based on the deep learning have been greatly improved compared with some traditional algorithms, it is still a big problem to clearly interpret the deep learning models. To address the issues, this work first introduces a popular deep learning model EEGNet and compares it with the traditional algorithm Filter-Bank Common Spatial Pattern (FBCSP). After that, this work considers that the 1-D convolution of EEGNet can be explained by a special Discrete Wavelet Transform (DWT), and the depthwise convolution of EEGNet is similar to the Common Spatial Pattern (CSP) algorithm. Therefore, this work improves the EEGNet by using the algorithm Temporary Constrained Sparse Group Lasso (TCSGL) to enhance its performance. The proposed model TSGL-EEGNet is tested on the BCI Competition IV 2a and BCI Competition III IIIa datasets that both are 4-classes classification MI tasks. The testing results show that the proposed model has achieved 78.96% (0.7194) average classification accuracy (kappa) on the dataset BCI Competition IV 2a, which are greater than EEGNet, C2CM, MB3DCNN, SS-MEMDBF and FBCSP, especially on insensitive subjects. The proposed model has also achieved 85.30% (0.8040) average classification accuracy (kappa) on the dataset BCI Competition III IIIa, which are greater than the EEGNet, MFTFS et al. At last, this work uses average-validation and stacking to further enhance the effect of the model. The 4-classes classification average accuracy rates reach 81.34% and 88.89%, and the kappas reach 0.7511 and 0.8519 on dataset BCI Competition IV 2a and BCI Competition III IIIa, respectively. Additionally, this work also uses the Grad-CAM to visualize the frequency and spatial features that are learned by the neural network.
A comparison between the hydrogen storage performances of untreated and heat-treated Li3N
was carried out using a volumetric method, temperature-programmed hydrogenation (TPH),
and scanning electron microscopy (SEM). It was found that, during the first hydrogenation, the
untreated Li3N could absorb fast (3 min), 5.5 wt % of hydrogen at 230 °C, and that 3200 min
was needed to absorb an additional 2.5 wt % of H2. In contrast, the Li3N preheated at 400 °C in
a vacuum for 4.5 h absorbed only 3 wt % in 3 min at 230 °C but required only 300 min to absorb
an additional 6.5 wt % of H2. Furthermore, during rehydrogenation after desorption, the
pretreated Li3N could quickly absorb (3 min) 5.2 wt % of hydrogen, and this performance did
not change with increasing hydrogenation−dehydrogenation cycle number. TPH and SEM
investigations indicated that the higher performance of the preheated Li3N was due to its
relatively large particle sizes.
A number of geodynamic models have been proposed for the southeastern margin of the Tibetan plateau and include a range of deformation processes. One unresolved issue is whether crustal and mantle flow occurs, and if it does, how flow contributes to the mass balance of the India-Asia collision. To address this question, new magnetotelluric data were collected and used to derive a three-dimensional electrical resistivity model of the crust and upper mantle beneath the Red River Fault (RRF) zone and adjacent areas. The most prominent features of the model are (1) a resistor in the upper-middle crust directly beneath the trace of the RRF; (2) a major change in upper mantle resistivity across the RRF; and (3) a significant conductor in the upper mantle northeast of the RRF, which extends upward into the crust, and which requires a melt fraction of up to 3%. The model suggests that the lower crustal conductors may be due to melt/fluids derived from the mantle, rather than from outward flow from Tibet. The most likely source of fluids and melts could be upwelling mantle flow related to the Hainan mantle plume. The change in resistivity across the RRF implies a change in lithospheric strength may explain the present-day localization of deformation and uplift at this location. The resistivity model may also give insights into the distribution of ore deposits in Ailao Shan, since many mineral deposits are derived from magmatic fluids generated in the mantle at the edges of regions of thick lithosphere. Plain Language Summary Continent-continent collisions are an important tectonics process that have formed mountain ranges such as the Himalaya and features such as the Tibetan Plateau. Some recent models have proposed that deformation in such regions is partly accommodated by ductile deformation processes such as crustal flow. They proposed that flow occurs locally overs tens of kilometers beneath the Himalaya and are supported by geological and geophysical evidence. Much larger crustal flow systems have been proposed to occur beneath the eastern margin of the Tibetan Plateau, where the collision has extruded large regions of the crust with upper crustal deformation focused on a system of major transform faults. This study investigates these models by using geophysical data to image the properties of rocks beneath the Red River Fault area of southeastern Tibet. The magnetotelluric method was used to map the resistivity of the rocks, which is a parameter sensitive to the presence of fluids, which in turn control the strength. Regions of low resistivity were mapped and inferred to be due regions of partial melt. These appear to originate in the mantle, and not be connected to similar features to the North. This implies that large-scale crustal flow may not occur in this region.
As a potential candidate for hydrogen storage, Li 3 N can absorb more than 9 wt % hydrogen. However, because of its incomplete dehydrogenation at a temperature of 280 °C, only about 5.5 wt % reversible hydrogen capacity could be reached. Although by increasing the temperature one can enhance dehydrogenation, this paper demonstrates that dehydrogenation of hydrogenated Li 3 N at the high temperature of 400 °C is followed by a very low (0.4 wt %) rehydrogenation capacity. Furthermore, scanning electron microscopy, Brunauer-Emmett-Teller surface area, and X-ray powder diffraction measurements have shown that both the sintering and the lattice structure change of Li 2 NH, which is a product of hydrogenation, might be responsible for such a major deactivation.
This study, via combined analysis of geophysical and geochemical data, reveals a lithospheric architecture characterized by crust-mantle decoupling and vertical heat-flow conduits that control orogenic gold mineralization in the Ailaoshan gold belt on the southeastern margin of Tibet. The mantle seismic tomography indicates that the crust-mantle decoupled deformation, defined from previous seismic anisotropy analysis was formed by upwelling and lateral flow of the asthenosphere driven by deep subduction of the Indian continent. Our MT and seismic images show both a vertical conductor across the Moho and high Vp/Vs anomalies both in the uppermost mantle and lowest crust, suggesting that crust-mantle decoupling promotes ponding of mantle-derived basic melts at the base of the crust via a heat-flow conduit. Noble gas isotope and halogen ratios of gold-related ore minerals indicate a mantle source of ore fluid, and a rapid decrease in Cl/F ratios of lamprophyres under P-T conditions of 1.2 Gpa and 1050 C suggests that the ore fluid was derived from degassing of the basic melts. Similar lithospheric architecture is recognized in other orogenic gold provinces, implying analogous formational controls.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.