A Railway Fastener Inspection Method Based on Abnormal Sample Generation

Regular fastener detection is necessary to ensure the safety of railways.However,the number of abnormal fasteners is significantly lower than the number of normal fasteners in real railways.Existing supervised inspectionmethods have insufficient detection ability in cases of imbalanced samples.To solve this problem,we propose an approach based on deep convolutional neural networks(DCNNs),which consists of three stages:fastener localization,abnormal fastener sample generation based on saliency detection,and fastener state inspection.First,a lightweight YOLOv5s is designed to achieve fast and precise localization of fastener regions.Then,the foreground clip region of a fastener image is extracted by the designed fastener saliency detection network(F-SDNet),combined with data augmentation to generate a large number of abnormal fastener samples and balance the number of abnormal and normal samples.Finally,a fastener inspection model called Fastener ResNet-8 is constructed by being trained with the augmented fastener dataset.Results show the effectiveness of our proposed method in solving the problem of sample imbalance in fastener detection.Qualitative and quantitative comparisons show that the proposed F-SDNet outperforms other state-of-the-art methods in clip region extraction,reaching MAE and max F-measure of 0.0215 and 0.9635,respectively.In addition,the FPS of the fastener state inspection model reached 86.2,and the average accuracy reached 98.7%on 614 augmented fastener test sets and 99.9%on 7505 real fastener datasets.

A Silicon Monoxide Lithium-Ion Battery Anode with Ultrahigh Areal Capacity

Silicon monoxide(SiO)is an attractive anode material for next-generation lithium-ion batteries for its ultra-high theoretical capacity of 2680 mAh g−1.The studies to date have been limited to electrodes with a rela-tively low mass loading(<3.5 mg cm^(−2)),which has seriously restricted the areal capacity and its potential in practical devices.Maximizing areal capacity with such high-capacity materials is critical for capitalizing their potential in practi-cal technologies.Herein,we report a monolithic three-dimensional(3D)large-sheet holey gra-phene framework/SiO(LHGF/SiO)composite for high-mass-loading electrode.By specifically using large-sheet holey graphene building blocks,we construct LHGF with super-elasticity and exceptional mechanical robustness,which is essential for accommodating the large volume change of SiO and ensuring the structure integrity even at ultrahigh mass loading.Additionally,the 3D porous graphene network structure in LHGF ensures excellent electron and ion transport.By systematically tailoring microstructure design,we show the LHGF/SiO anode with a mass loading of 44 mg cm^(−2)delivers a high areal capacity of 35.4 mAh cm^(−2)at a current of 8.8 mA cm^(−2)and retains a capacity of 10.6 mAh cm^(−2)at 17.6 mA cm^(−2),greatly exceeding those of the state-of-the-art commercial or research devices.Furthermore,we show an LHGF/SiO anode with an ultra-high mass loading of 94 mg cm^(−2)delivers an unprecedented areal capacity up to 140.8 mAh cm^(−2).The achievement of such high areal capacities marks a critical step toward realizing the full potential of high-capacity alloy-type electrode materials in practical lithium-ion batteries.

Prelithiation strategies for silicon-based anode in high energy density lithium-ion battery

Green energy storage devices play vital roles in reducing fossil fuel emissions and achieving carbon neutrality by 2050.Growing markets for portable electronics and electric vehicles create tremendous demand for advanced lithium-ion batteries(LIBs)with high power and energy density,and novel electrode material with high capacity and energy density is one of the keys to next-generation LIBs.Silicon-based materials,with high specific capacity,abundant natural resources,high-level safety and environmental friendliness,are quite promising alternative anode materials.However,significant volume expansion and redundant side reactions with electrolytes lead to active lithium loss and decreased coulombic efficiency(CE)of silicon-based material,which hinders the commercial application of silicon-based anode.Prelithiation,preembedding extra lithium ions in the electrodes,is a promising approach to replenish the lithium loss during cycling.Recent progress on prelithiation strategies for silicon-based anode,including electrochemical method,chemical method,direct contact method,and active material method,and their practical potentials are reviewed and prospected here.The development of advanced Si-based material and prelithiation technologies is expected to provide promising approaches for the large-scale application of silicon-based materials.

A Detection Method of Bolts on Axlebox Cover Based on Cascade Deep Convolutional Neural Network

This paper proposes a cascade deep convolutional neural network to address the loosening detection problem of bolts on axlebox covers.Firstly,an SSD network based on ResNet50 and CBAM module by improving bolt image features is proposed for locating bolts on axlebox covers.And then,theA2-PFN is proposed according to the slender features of the marker lines for extracting more accurate marker lines regions of the bolts.Finally,a rectangular approximationmethod is proposed to regularize themarker line regions asaway tocalculate the angle of themarker line and plot all the angle values into an angle table,according to which the criteria of the angle table can determine whether the bolt with the marker line is in danger of loosening.Meanwhile,our improved algorithm is compared with the pre-improved algorithmin the object localization stage.The results show that our proposed method has a significant improvement in both detection accuracy and detection speed,where ourmAP(IoU=0.75)reaches 0.77 and fps reaches 16.6.And in the saliency detection stage,after qualitative comparison and quantitative comparison,our method significantly outperforms other state-of-the-art methods,where our MAE reaches 0.092,F-measure reaches 0.948 and AUC reaches 0.943.Ultimately,according to the angle table,out of 676 bolt samples,a total of 60 bolts are loose,69 bolts are at risk of loosening,and 547 bolts are tightened.

Design principle of single-atom catalysts for sulfur reduction reaction—interplay between coordination patterns and transition metals

The polysulfide shuttling effect is the primary bottleneck restricting the industrial application of Li-S batteries,and the electrocatalytic sulfur reduction reaction(SRR)has emerged as an effective solution.Carbon-based singleatom catalysts(SACs),which promotes SRR,show great potential in inhibiting the shuttling effect of polysulfides.Meanwhile,the optimization and rational design of such catalysts requires a deep understanding to the fundamental SRR mechanism and remains highly nontrivial.In this work,we construct a comprehensive database of carbon-based SACs,covering different coordination patterns,heteroatoms,and transition metals.The SRR activities are determined using density functional theory calculations,revealing a synergistic effect between the p orbital of the heteroatom and the d orbital of the transition metal.This interplay underscores the critical importance of the coordination environment for SRR under the ortho-P_(2)C_(2)structure.Regardless of the transition metal type,the ortho-P_(2)C_(2)coordination pattern significantly enhances the SRR performance of SACs,surpassing the widely reported N_(3)C_(1)and N_(4)coordinated graphene-based SACs.Furthermore,heteroatoms with ortho-P_(2)C_(2)may exhibit SRR activity.In a word,by using this comprehensive dataset and data-driven framework,we propose a promising novel class of coordination structure(ortho-P_(2)C_(2)structure)and neglected design principle.