检索结果-南通市图书馆

Defect detection and repair algorithm for structures generated by topology optimization based on 3D hierarchical fully convolutional network

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Biomimetic Intelligence & Robotics 2024年第2期4卷 24-31页

作者： Zhiyu wan Hai Lan Sichao Lin Houde Dai School of Advanced Manufacturing Fuzhou UniversityJinjiang 362251China Quanzhou Institute of Equipment Manufacturing Haixi InstitutesChinese Academy of SciencesJinjiang 362216China Fujian Key Laboratory of Special Intelligent Equipment Safety Measurement and Control Fujian Special Equipment Inspection and Research InstituteFuzhou 350008China

Customized 3D-printed structural parts are widely used in surgical *** satisfy the mechanical properties and kinematic functions of these structural parts,a topology optimization technique is adopted to obtain the optimal structural layout while meeting the constraints and ***,topology optimization currently faces some practical challenges that must be addressed,such as ensuring that structures do not have significant defects when converted to additive manufacturing *** address this problem,we designed a 3D hierarchical fully convolutional network(FCN)to predict the precise position of the defective *** on the prediction results,an effective repair strategy is adopted to repair the defective structure.A series of experiments is conducted to demonstrate the effectiveness of our *** to the 2D fully convolutional network and the rule-based detection method,our approach can accurately capture most defect structures and achieve 89.88%precision and 95.59%***,we investigate the impact of different ways to increase the receptive field of our model,as well as the trade-off between different defect-repairing *** results of the experiment demonstrate that the hierarchical structure,which increases the receptive field,can substantially improve the defect detection *** the best of our knowledge,this paper is the first to investigate 3D defect prediction and repair for topology optimization in conjunction with deep learning algorithms,providing practical tools and new perspectives for the subsequent development of topology optimization techniques.

关键词： Topology optimization Additive manufacturing Deep learning 3D semantic segmentation Defect detection

Significant enhancement in local thermal conductivity of erythritol at interface with nanoparticles due to their interaction

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nano Research 2024年第8期17卷 7726-7736页

作者： Ning Cao Yanhui Feng Kening Yan Zihan Liu Jiawei wan Lin Qiu School of Energy and Environmental Engineering University of Science and Technology BeijingBeijing 100083China State Key Laboratory of Biochemical Engineering Key Laboratory of Biopharmaceutical Preparation and DeliveryInstitute of Process EngineeringChinese Academy of SciencesBeijing 100190China

When nano-fillers are used to enhance the thermal conductivity of organic phase change materials(PCMs),the naturally formed interface is considered to hinder thermal transport of the composite ***,the effect of the interface on the thermal properties of surrounding PCM has not been fully *** this paper,three composite PCMs(Ery@SiC,Ery@SiO_(2) and Ery@Si_(3)N_(4))were prepared by melt-blending *** local thermal conductivity and reduced Young’s modulus(E^(*))of the erythritol at the interface and far away from the interface in the composite PCMs were simultaneously measured by scanning thermal microscopy(SThM).The results revealed significant enhancement in local thermal conductivity of erythritol at the interface and its obvious positive correlation with E^(*).For different composite PCMs,molecular dynamics(MD)simulations suggested that the increase in intrinsic thermal conductivity and E^(*)of erythritol is attributed to the increase in interaction energy between erythritol and nanoparticles,as more erythritol phonon vibrations transform from localized mode to delocalized mode and erythritol has a higher density at the *** findings will provide new ideas for the design of PCM for energy storage.

关键词： scanning thermal microscopy phase change materials thermal conductivity reduced Young’s modulus interaction energy phonon participation ratio

Attention-based network embedding with higher-order weights and node attributes

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2024年第2期9卷 440-451页

作者： Xian Mo Binyuan wan Rui Tang Junkai Ding Guangdi Liu School of Information Engineering Ningxia UniversityYinchuanChina School of Cyber Science and Engineering Sichuan UniversityChengduChina Computer Science and Information Technology University of MalayaKuala LumpurMalaysia Academy of Medical Sciences Zhengzhou UniversityZhengzhouChina

Network embedding aspires to learn a low-dimensional vector of each node in networks,which can apply to diverse data mining *** real-life,many networks include rich attributes and temporal ***,most existing embedding approaches ignore either temporal information or network attributes.A self-attention based architecture using higher-order weights and node attributes for both static and temporal attributed network embedding is presented in this article.A random walk sampling algorithm based on higher-order weights and node attributes to capture network topological features is *** static attributed networks,the algorithm incorporates first-order to k-order weights,and node attribute similarities into one weighted graph to preserve topological features of *** temporal attribute networks,the algorithm incorporates previous snapshots of networks containing first-order to k-order weights,and nodes attribute similarities into one weighted *** addition,the algorithm utilises a damping factor to ensure that the more recent snapshots allocate a greater *** features are then incorporated into topological ***,the authors adopt the most advanced architecture,Self-Attention Networks,to learn node *** results on node classification of static attributed networks and link prediction of temporal attributed networks reveal that our proposed approach is competitive against diverse state-of-the-art baseline approaches.

关键词： data mining deep neural networks social network

Identification and Molecular Characterization of the Alkaloid Biosynthesis Gene Family in Dendrobium catenatum

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Phyton-International Journal of Experimental Botany 2024年第1期93卷 81-96页

作者： Liping Yang Xin wan Runyang Zhou Yingdan Yuan College of Horticulture and Landscape Architecture Yangzhou UniversityYangzhouChina Jiangsu Academy of Forestry NanjingChina Jiangsu Yangzhou Urban Forest Ecosystem National Observation and Research Station YangzhouChina

As one of the main active components of Dendrobium catenatum, alkaloids have high medicinal value. The physicochemicalproperties, conserved domains and motifs, phylogenetic analysis, and cis-acting elements of the genefamily members in the alkaloid biosynthesis pathway of D. catenatum were analyzed by bioinformatics, and theexpression of the genes in different years and tissues was analyzed by qRT-PCR. There are 16 gene families,including 25 genes, in the D. catenatum alkaloid biosynthesis pathway. The analysis of conserved domains andmotifs showed that the types, quantities, and orders of domains and motifs were similar among members ofthe same family, but there were significant differences among families. Phylogenetic analysis indicated that thegene family members showed some evolutionary conservation. Cis-acting element analysis revealed that therewere a large number of light-responsive elements and MYB (v-myb avian myeloblastosis viral oncogene homolog)-related elements in these genes. qRT-PCR showed that expressions of gene family members involved in alkaloidsynthesis were different in different years and tissues of D. catenatum. This study provides a theoretical basisfor further exploration of the regulatory mechanisms of these genes in the alkaloid biosynthesis of D. catenatum.

关键词： Dendrobium catenatum gene family alkaloid biosynthesis

HCSP-Net:A Novel Model of Age-Related Macular Degeneration Classification Based on Color Fundus Photography

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第4期79卷 391-407页

作者： Cheng wan Jiani Zhao Xiangqian Hong Weihua Yang Shaochong Zhang College of Electronic and Information Engineering/College of Integrated Circuits Nanjing University of Aeronautics and AstronauticsNanjing211106China Shenzhen Eye Institute Shenzhen Eye HospitalJinan UniversityShenzhen518040China

Age-related macular degeneration(AMD)ranks third among the most common causes of *** the most conventional and direct method for identifying AMD,color fundus photography has become prominent owing to its consistency,ease of use,and good quality in extensive clinical *** this study,a convolutional neural network(CSPDarknet53)was combined with a transformer to construct a new hybrid model,*** hybrid model was employed to tri-classify color fundus photography into the normal macula(NM),dry macular degeneration(DMD),and wet macular degeneration(WMD)based on clinical classification manifestations,thus identifying and resolving AMD as early as possible with color fundus *** further enhance the performance of this model,grouped convolution was introduced in this study without significantly increasing the number of ***-Net was validated using an independent test *** average precision of HCSPNet in the diagnosis of AMD was 99.2%,the recall rate was 98.2%,the F1-Score was 98.7%,the PPV(positive predictive value)was 99.2%,and the NPV(negative predictive value)was 99.6%.Moreover,a knowledge distillation approach was also adopted to develop a lightweight student network(SCSP-Net).The experimental results revealed a noteworthy enhancement in the accuracy of SCSP-Net,rising from 94%to 97%,while remarkably reducing the parameter count to a quarter of *** attribute positions SCSP-Net as a highly suitable candidate for the deployment of resource-constrained devices,which may provide ophthalmologists with an efficient tool for diagnosing AMD.

关键词： Computer-aided diagnosis deep learning age-related macular degeneration transformer

The combined effect of bromadiolone and ivermectin (iBr) in controlling both rodents and their fleas

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Integrative Zoology 2024年第1期19卷 156-164页

作者： Ming LIU Xinrong wan Wei LIU Xingan MA Zhibin ZHANG State Key Laboratory of Integrated Pest Management Institute of ZoologyChinese Academy of SciencesBeijingChina CAS Center for Excellence in Biotic Interactions University of Chinese Academy of SciencesBeijingChina Inner Mongolia Minzu University TongliaoInner MongoliaChina International Society of Zoological Sciences BeijingChina

Rodent pests not only cause severe agricultural loss but also spread zoonotic pathogens to human ***-agulant rodenticides are widely used to decrease the population densities of rodents but often lead to the spillover of ectoparasites becausefleas and ticks may gather on surviving ***,it is necessary to killfleas and ticks before culling rodents to minimize the risk of pathogen *** this study,we used a mixture of ivermectin(an antiparasitic drug)and bromadiolone(an anticoagulant rodenticide)to control both rodent andflea/tick *** found that in a laboratory test,0.01%ivermectin bait was not lethal for greater long-tailed hamsters after 7 days of treatment,while 0.1%ivermectin bait was lethal for approximately 33%of treated *** afield test,bait containing 0.001%,0.005%,0.01%,and 0.05%ivermectin decreased the number offleas per vole of Brandt’s voles to 0.42,0.22,0.12,and 0.2,respectively,compared with 0.77 in the control group,indicating that 0.01%ivermectin bait performed best in *** another laboratory test,bait containing a 0.01%ivermectin and 0.005%bromadiolone mixture caused the death of all voles within 6–14 days after the intake of the *** thefield test,the bait containing 0.01%ivermectin and 0.005%bromadiolone reduced the average number offleas per vole to 0.35,which was significantly lower than the 0.77 of the control *** results indicate that a 0.01%ivermectin and 0.005%bromadiolone mixture could be used to control both rodents andfleas to minimize the spillover risk of disease transmission when using traditional rodenticides.

关键词： bromadiolone flea iBr ivermectin rodent synergistic control

Programmable meta-holography dynamics enabled by grating-modulation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Photonics Research 2024年第7期12卷 1522-1526页

作者： RUNLONG RAO SHUAI wan ZHE LI YANGYANG SHI ZHONGYANG LI Electronic Information School Wuhan UniversityWuhan 430072China Wuhan Institute of Quantum Technology Wuhan 430206China School of Microelectronics Wuhan UniversityWuhan 430072China Suzhou Institute of Wuhan University Suzhou 215123China

Towards next-generation intelligent display devices, it is urgent to find a cheap and convenient dynamic optical control method. Conventional gratings are widely used as cheap diffractive elements due to their effective optical control capabilities. However, they are limited within multi-function or complex optical modulation due to the lack of accurate control of the amplitude/phase at pixel-level. Here, a metasurface-assisted grating-modulation system is originally proposed to achieve switchable multi-fold meta-holographic dynamics. By incorporating metasurfaces with traditional gratings and tuning their relative coupling positions, the modulation system gains the optical modulation capability to realize complex optical functionalities. Specifically, by varying the grating periods/positions relative to the metasurface, 2–8 holographic image channels are programmed to be dynamically exhibited and switched. The proposed metasurface-assisted grating-modulation approach enjoys cost-effective convenience, strong encoding freedom, and facile operation, which promises programmable optical displays,optical sensors, optical information encryption/storage, etc.

关键词： tuning switched holographic

Graph Convolutional Networks Embedding Textual Structure Information for Relation Extraction

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第5期79卷 3299-3314页

作者： Chuyuan Wei Jinzhe Li Zhiyuan wang Shanshan wan Maozu Guo School of Electrical and Information Engineering Beijing University of Civil Engineering and ArchitectureBeijing102616China

Deep neural network-based relational extraction research has made significant progress in recent years,andit provides data support for many natural language processing downstream tasks such as building knowledgegraph,sentiment analysis and question-answering ***,previous studies ignored much unusedstructural information in sentences that could enhance the performance of the relation extraction ***,most existing dependency-based models utilize self-attention to distinguish the importance of context,whichhardly deals withmultiple-structure *** efficiently leverage multiple structure information,this paperproposes a dynamic structure attention mechanism model based on textual structure information,which deeplyintegrates word embedding,named entity recognition labels,part of speech,dependency tree and dependency typeinto a graph convolutional ***,our model extracts text features of different structures from theinput *** Structure information Graph Convolutional Networks employs the dynamic structureattention mechanism to learn multi-structure attention,effectively distinguishing important contextual features invarious structural *** addition,multi-structure weights are carefully designed as amergingmechanismin the different structure attention to dynamically adjust the final *** paper combines these featuresand trains a graph convolutional network for relation *** experiment on supervised relation extractiondatasets including SemEval 2010 Task 8,TACRED,TACREV,and Re-TACED,the result significantly outperformsthe previous.

关键词： Relation extraction graph convolutional neural networks dependency tree dynamic structure attention

Insights into the role of oxygen-containing functional groups on carbon surface in water–electricity generation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nano Research 2024年第7期17卷 6645-6653页

作者： wan Xue Zongbin Zhao Honghui Bi Bolun Zhang Xuzhen wang Jieshan Qiu State Key Laboratory of Fine Chemicals Liaoning Key Laboratory for Energy Materials and Chemical EngineeringSchool of Chemical EngineeringDalian University of TechnologyDalian 116024China State Key Laboratory of Fine Chemicals Liaoning Key Laboratory for Energy Materials and Chemical EngineeringSchool of ChemistryDalian University of TechnologyDalian 116024China State Key Laboratory of Chemical Resource Engineering College of Chemical EngineeringBeijing University of Chemical TechnologyBeijing 100029China

A deep understanding of the electricity generation mechanism from the interaction between water molecules and carbon material surfaces is attractive for next-generation water-based energy conversion and storage ***,an asymmetric generator was assembled based on functionalized carbon nanotubes films to investigate the relative contribution from various oxygen functional groups on carbon surface to the water-electrical *** and calculations demonstrate that the electricity mainly originates from the water molecule adsorption by carboxyl groups and dissociation of functional groups on carbon surface,which leads to the formation of electrical double layers at *** device allows the electricity generation with a variety of water sources,such as deionized water,tap water,as well as *** particular,the generator based on carboxyl carbon nanotubes can induce a voltage of over 200 mV spontaneously in natural seawater with the power density of about 0.11 mW·g^(−1).High voltages can be achieved easily through the series-connection strategy to power electronic products such as a liquid crystal *** work reveals the dominant role of carboxyl groups in carbon-based water–electricity conversion and is expected to offer inspiration for the preparation of carbon materials with high electrical performance.

关键词： electricity generation water adsorption dissociation carboxyl groups carbon nanotubes