作者机构：
[Zhang, Cong; Miao, Qing; Zhang, C; Liu, Ke; Zheng, Ziang; Chen, Dachang] Wuhan Polytech Univ, Sch Elect & Elect Engn, Wuhan 430023, Peoples R China.;[Luo, Yi] Chinese Acad Sci, Inst Elect Engn, Beijing Int S&T Cooperat Base Plasma Sci & Energy, Beijing 100190, Peoples R China.;[Xiao, Song] Wuhan Univ, Sch Elect Engn & Automat, Wuhan, Peoples R China.;[Xiao, Beibei] Jiangsu Univ Sci & Technol, Sch Energy & Power Engn, Zhenjiang, Peoples R China.
通讯机构：
[Luo, Y ] C;[Zhang, C ] W;Wuhan Polytech Univ, Sch Elect & Elect Engn, Wuhan 430023, Peoples R China.;Chinese Acad Sci, Inst Elect Engn, Beijing Int S&T Cooperat Base Plasma Sci & Energy, Beijing 100190, Peoples R China.
摘要：
Heptafluoroisobutyronitrile (C 4 F 7 N) is being considered as a promising alternative to the greenhouse gas SF 6 in the electrical industry. However, its biotoxicity necessitates the development of gas sensing technology to detect leaked C 4 F 7 N. A combination of density functional theory and experiments was employed to evaluate the adsorption and sensing performance of different metal-phthalocyanines as potential sensing materials for C 4 F 7 N detection. The study included exploring adsorption configurations with adsorption energies, electron transfer, and adsorption distance, as well as comparisons of electronic properties, electron distribution, and density of states (DOS) among the MPcs. Furthermore, gas sensing experiments were conducted using different MPcs to detect 25–100 ppm C 4 F 7 N. The results revealed that Mn-Pc, Fe-Pc, Co-Pc, and Zn-Pc exhibited considerable chemical interactions, while Ni-Pc and Cu-Pc showed weaker adsorption strength. These findings were further elucidated based on the electron density and DOS of atomic orbitals. Moreover, gas sensing experiments indicated that Co-Pc demonstrated a higher response compared to Fe-Pc at the same concentration of C 4 F 7 N. Overall, the theoretical and experimental insights offer valuable guidance for C 4 F 7 N detection and provide a systematic approach to screen and explore organometallic polymer-based gas sensors applicable in various fields.
摘要：
Foreign fibers directly impact the quality of raw cotton, affecting the prices of textile products and the economic efficiency of cotton textile enterprises. The accurate differentiation and labeling of foreign fibers require domain-specific knowledge, and labeling scattered cotton foreign fibers in images consumes substantial time and labor costs. In this study, we propose a semi-supervised foreign fiber detection approach that uses unlabeled image information and a small amount of labeled data for model training. Our proposed method, Efficient YOLOv5-cotton, introduces CBAM to address the issue of the missed detection and false detection of small-sized cotton foreign fibers against complex backgrounds. Second, the algorithm designs a multiscale feature information extraction network, SPPFCSPC, which improves its ability to generalize to fibers of different shapes. Lastly, to reduce the increased network parameters and computational complexity introduced by the SPPFCSPC module, we replace the C3 layer with the C3Ghost module. We evaluate Efficient YOLOv5 for detecting various types of foreign fibers. The results demonstrate that the improved Efficient YOLOv5-cotton achieves a 1.6% increase in mAP@0.5 (mean average precision) compared with the original Efficient YOLOv5 and reduces model parameters by 10% compared to the original Efficient YOLOv5 with SPPFCSPC. Our experiments show that our proposed method enhances the accuracy of foreign fiber detection using Efficient YOLOv5-cotton and considers the trade-off between the model size and computational cost.
通讯机构：
[Zeng, S ; Liu, WH ] W;Wuhan Polytech Univ, Sch Elect & Elect Engn, Wuhan 430023, Peoples R China.;Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Peoples R China.
关键词：
fresh-cut flower sorting system;RGBD images;multi-task and multi-dimension-You Only Look Once
摘要：
As the quality of life rises, the demand for flowers has increased significantly, leading to higher expectations for flower sorting system efficiency and speed. This paper presents a real-time, high-precision end-to-end method, which can complete three key tasks in the sorting system: flower localization, flower classification, and flower grading. In order to improve the challenging maturity detection, red–green–blue depth (RGBD) images were captured. The multi-task and multi-dimension-You Only Look Once (MTMD-YOLO) network was proposed to complete these three tasks in an end-to-end manner. The feature fusion was simplified to increase training speed, and the detection head and non-maximum suppression (NMS) were optimized for the dataset. This optimization allowed the loss function for the grading task to be added to train each task separately. The results showed that the use of RGBD and multi-task improved by 3.63% and 1.87% of mean average precision (mAP) on flower grading task, respectively. The final mAP of the flower classification and grading task reached 98.19% and 97.81%, respectively. The method also achieved real-time speed on embedded Jetson Orin NX, with 37 frames per second (FPS). This method provided essential technical support to determine the automatic flower picking times, in combination with a picking robot.
关键词：
RBFNN control structure;compliant motion constraint;integrated framework;iterative learning;robot-assisted rehabilitation
摘要：
INTRODUCTION: The robot-assistive technique has been widely developed in the field of neurorehabilitation for enhancement of neuroplasticity, muscle activity, and training positivity. To improve the reliability and feasibility in this patient-robot interactive context, motion constraint methods and adaptive assistance strategies have been developed to guarantee the movement safety and promote the training effectiveness based on the user's movement information. Unfortunately, few works focus on customizing quantitative and appropriate workspace for each subject in passive/active training mode, and how to provide the precise assistance by considering movement constraints to improve human active participation should be further delved as well. METHODS: This study proposes an integrated framework for robot-assisted upper-limb training. A human kinematic upper-limb model is built to achieve a quantitative human-robot interactive workspace, and an iterative learning-based repulsive force field is developed to balance the compliant degrees of movement freedom and constraint. On this basis, a radial basis function neural network (RBFNN)-based control structure is further explored to obtain appropriate robotic assistance. The proposed strategy was preliminarily validated for bilateral upper-limb training with an end-effector-based robotic system. RESULTS: Experiments on healthy subjects are enrolled to validate the safety and feasibility of the proposed framework. The results show that the framework is capable of providing personalized movement workspace to guarantee safe and natural motion, and the RBFNN-based control structure can rapidly converge to the appropriate robotic assistance for individuals to efficiently complete various training tasks. DISCUSSION: The integrated framework has the potential to improve outcomes in personalized movement constraint and optimized robotic assistance. Future studies are necessary to involve clinical application with a larger sample size of patients.
摘要：
While highly sensitive elastic strain sensors have been widely investigated, the time-dependent stress sensitivity of viscoelastic MWCNT/polyethylene nanocomposite stress sensors remains to be explored. In this paper, we develop an electromechanically coupled homogenization scheme to reveal the time-dependent stress sensing performances of viscoelastic MWCNT/polyethylene nanocomposite sensors. In the time-dependent context, the complex moduli and electrical conductivity are selected as the dual homogenization parameters. The time-dependent stress sensitivity is illustrated through the viscoelastic imperfect interface connection and stress-induced tunneling distance. The predicted stress sensing capacities of viscoelastic MWCNT/polyethylene nanocomposite stress sensors are shown to be consistent with the experiments under the constant stress loading. It reveals that the stress sensitivity factor increases with the loading time under constant stress. The optimal MWCNT aspect ratio for high sensing capacities exhibits an increasing trend regarding the MWCNT volume fraction. The uncovered sensing characteristics can provide microstructural design guidance in high-performance nanocomposite stress sensors.
摘要：
Under the global trend of green development, the research and development of environmentally friendly gas- insulated electrical equipment and the associated technical issues in the power systems industry have become important research topics. This study investigates the monitoring technology for decomposition gases (CO, C2F6, and COF2) under insulation defect conditions in environmentally friendly GIS equipment based on the green insulating gas mixture C4F7N/CO2. Initially, chromium clusters (Crn, n = 1-4) were used to modify onedimensional boron nitride nanotubes (BNNTs), describing the changes in semiconductor electronic properties and magnetism at the microscopic interface. Interestingly, the electronic properties of the system changed significantly after modification, with increased electron mobility and enhanced conductivity. Subsequently, adsorption tests were conducted on three potential target gases, revealing that Cr-BNNT exhibited excellent adsorption for COF2, with an adsorption energy reaching-9.555 eV. This indicates its potential as a clean material for addressing COF2 (toxic) and CO leaks. Lastly, and most importantly, in the assessment of sensing performance, it has been discovered that Cr-BNNT holds promise as a gas sensor for C2F6. Furthermore, to meet diverse sensing requirements and environmental conditions, Cr3-BNNT and Cr4-BNNT have demonstrated potential as sensor array materials for the detection of CO (300 K, 0.145 mu s) and COF2 (500 K, 3 min) gas concentrations, respectively. Our study contributes to the advancement of novel eco-friendly insulating gases and explores the use of new materials for monitoring the insulation performance and condition assessment of environmentally friendly GIS equipment.
摘要：
目的:改善现有水果识别与分级方法依赖于人工操作和复杂设备的情况.方法:提出了一种轻量化模型YOLO-FFD(YOLO with fruit and freshen detection),该模型以YOLOv5框架为基础,基于深度可分离卷积和GELU激活函数设计轻量化模块LightweightC3作为主干特征提取网络的基本单元,减少模型参数量和计算量,加快模型的收敛速度;使用大内核深度可分离卷积模块EnhancedC3改进原模型的颈部,抑制信息丢失并增强模型的特征融合能力,提高模型的检测精度;采用GSConv代替特征融合网络中的普通卷积,使模型进一步轻量化.结果:提出模型的平均精度均值达到了 96.12％,在RTX 3090上速度为172帧/s,在嵌入式设备Jetson TX2上速度为20帧/s.相比于原始YOLOv5模型,平均精度均值提高了 2.21％,计算量减少了 26％,在RTX 3090和Jetson TX2上的速度分别提高了 2倍和1倍.结论:YOLO-FFD模型能够满足识别水果品种和新鲜度的需求,且在复杂场景下错检、漏检情况均有改善.
摘要：
We propose a theoretical scheme to realize a two-dimensional (2D) diffraction grating in a four-level inverted-Y-type atomic system coupled by a standing-wave (SW) field and a Laguerre–Gaussian (LG) vortex field. Owing to asymmetric spatial modulation of the LG vortex field, the incident probe field can be lopsidedly diffracted into four domains and an asymmetric 2D electromagnetically induced grating is created. By adjusting the detunings of the probe field and the LG vortex field, the intensities of the LG vortex field and the coherent SW field, as well as the interaction length, the diffraction properties and efficiency, can be effectively manipulated. In addition, the effect of the azimuthal parameter on the Fraunhofer diffraction of the probe field is also discussed. This asymmetric 2D diffraction grating scheme may provide a versatile platform for designing quantum devices that require asymmetric light transmission.
通讯机构：
[Qi, X ; Liu, HT; Huang, ZY] X;[Liu, HT ] W;Wuhan Polytech Univ, Sch Elect & Elect Engn, Wuhan 430023, Peoples R China.;Xiangtan Univ, Sch Phys & Optoelect, Hunan Key Lab Micronano Energy Mat & Devices, Xiangtan 411105, Peoples R China.
摘要：
In recent years, doping engineering, which is widely studied in theoretical and experimental research, is an effective means to regulate the crystal structure and physical properties of two-dimensional materials and expand their application potential. Based on different types of element dopings, different 2D materials show different properties and applications. In this paper, the characteristics and performance of rich layered 2D materials under different types of doped elements are comprehensively reviewed. Firstly, 2D materials are classified according to their crystal structures. Secondly, conventional experimental methods of charge doping and heterogeneous atom substitution doping are summarized. Finally, on the basis of various theoretical research results, the properties of several typical 2D material representatives under charge doping and different kinds of atom substitution doping as well as the inspiration and expansion of doping systems for the development of related fields are discussed. Through this review, researchers can fully understand and grasp the regulation rules of different doping engineering on the properties of layered 2D materials with different crystal structures. It provides theoretical guidance for further improving and optimizing the physical properties of 2D materials, improving and enriching the relevant experimental research and device application development. Doping engineering, including doping non-metallic atoms, alkali metal atoms, transition metal atoms and other metal atoms can be widely used in a variety of different structures of graphene-like novel 2D materials.
通讯机构：
[Xu, Z ] W;Wuhan Polytech Univ, Sch Elect & Elect Engn, Wuhan, Peoples R China.
关键词：
data fusion;ring buffer;sliding window;support degree function;wireless sensor networks
摘要：
In wireless sensor network (WSN) monitoring systems, redundant data from sluggish environmental changes and overlapping sensing ranges can increase the volume of data sent by nodes, degrade the efficiency of information collection, and lead to the death of sensor nodes. To reduce the energy consumption of sensor nodes and prolong the life of WSNs, this study proposes a dual layer intracluster data fusion scheme based on ring buffer. To reduce redundant data and temporary anomalous data while guaranteeing the temporal coherence of data, the source nodes employ a binarized similarity function and sliding quartile detection based on the ring buffer. Based on the improved support degree function of weighted Pearson distance, the cluster head node performs a weighted fusion on the data received from the source nodes. Experimental results reveal that the scheme proposed in this study has clear advantages in three aspects: the number of remaining nodes, residual energy, and the number of packets transmitted. The data fusion of the proposed scheme is confined to the data fusion of the same attribute environment parameters.
通讯机构：
[Yang, L ] W;Wuhan Polytech Univ, Coll Mech Engn, Wuhan 430048, Hubei, Peoples R China.
关键词：
Brown rice;Bran layer;Micro-structure;Moderate processing
摘要：
Moderate processing can improve the edible rice grain quality and nutrition. The grain structural-removal behavior was researched based on the macro and micro-structure analysis. Grain bran layer 3D profiling combined with iodine solution staining is applied for geometrical-structural analysis. Results show that rice grain can be divided into five structural region based on bran layer removal capacity, lateral > ventral > ventral groove > dorsal > dorsal groove. The relationship between grain geometric parameters and collision possibility is in-depth analyzed, results indicated large grain surface curvature leads to grain surface layer hard removal capacity. The region's bran layer thickness distribution affects its removal behavior and obey its removal capacity. The regional bran layer thickness order: dorsal(59.56 μm) > dorsal groove(50.62 μm) > ventral(43.91 μm) > ventral groove(42.83 μm) > lateral(37.08 μm). The bran layer micro-morphology impacts removal behavior, dorsal region with the groove structure, the bottom surface layer in grain groove structure can be removed when groove structure is damaged. The ventral groove depth is less than dorsal groove, showing better removal ability. The bran layer color in L*A*B*space shows strong correlation with remaining layer thickness, reflecting grain milling degree. Combined effects of grain geometry-parameters, bran layer thickness and micro-structure lead the final bran layer removal behavior. This study provides theoretical and practical basis for grain moderate processing.
Moderate processing can improve the edible rice grain quality and nutrition. The grain structural-removal behavior was researched based on the macro and micro-structure analysis. Grain bran layer 3D profiling combined with iodine solution staining is applied for geometrical-structural analysis. Results show that rice grain can be divided into five structural region based on bran layer removal capacity, lateral > ventral > ventral groove > dorsal > dorsal groove. The relationship between grain geometric parameters and collision possibility is in-depth analyzed, results indicated large grain surface curvature leads to grain surface layer hard removal capacity. The region's bran layer thickness distribution affects its removal behavior and obey its removal capacity. The regional bran layer thickness order: dorsal(59.56 μm) > dorsal groove(50.62 μm) > ventral(43.91 μm) > ventral groove(42.83 μm) > lateral(37.08 μm). The bran layer micro-morphology impacts removal behavior, dorsal region with the groove structure, the bottom surface layer in grain groove structure can be removed when groove structure is damaged. The ventral groove depth is less than dorsal groove, showing better removal ability. The bran layer color in L*A*B*space shows strong correlation with remaining layer thickness, reflecting grain milling degree. Combined effects of grain geometry-parameters, bran layer thickness and micro-structure lead the final bran layer removal behavior. This study provides theoretical and practical basis for grain moderate processing.
摘要：
In order to address the issues of low accuracy and limited generalization ability in current facial expression recognition, a facial expression recognition model based on VGG16 and attention mechanism is proposed.The proposed model is based on VGG16 and embeds Switchable Normalization layers and SE channel attention mechanism modules into the network,enabling the model to dynamically select weights for different normalized operation and channels.Additionally,the dataset is subjected to various image preprocessing techniques such as rotation and horizontal flipping.Experimental results demonstrate that the proposed model achieves an impressive accuracy rate of 73.22% on the FER2013 dataset,which outperforms many existing mainstream methods and achieves good performance in facial expression recognition.
