检索结果-南通市图书馆

A deep neural network based surrogate model for damage identification in full-scale structures with incomplete noisy measurements

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Structural and Civil Engineering 2024年第3期18卷 393-410页

作者： tram BUI-ngoc Duy-Khuong LY Tam T TRUONG Chanachai THONGCHOM T.nguyen-THOI Laboratory for Computational Mechanics Institute for Computational Science and Artificial IntelligenceVan Lang UniversityHo Chi Minh City 70000Vietnam Faculty of Mechanical-Electrical and Computer Engineering School of TechnologyVan Lang UniversityHo Chi Minh City 70000Vietnam Faculty of Civil Engineering School of TechnologyVan Lang UniversityHo Chi Minh City 70000Vietnam Department of Computer Science Aarhus UniversityAarhus 8000Denmark Thammasat University research unit in structural and foundation engineering Department of Civil EngineeringThammasat UniversityPathumthani 12120Thailand Laboratory for Applied and Industrial Mathematics Institute for Computational Science and Artificial IntelligenceVan Lang UniversityHo Chi Minh City 70000Vietnam Thammasat School of Engineering Thammasat UniversityPathumthani 12120Thailand

The paper introduces a novel approach for detecting structural damage in full-scale structures using surrogate models generated from incomplete modal data and deep neural networks(DNNs).A significant challenge in this field is the limited availability of measurement data for full-scale structures,which is addressed in this paper by generating data sets using a reduced finite element(FE)model constructed by SAP2000 software and the MATLAB programming *** surrogate models are trained using response data obtained from the monitored structure through a limited number of measurement *** proposed approach involves training a single surrogate model that can quickly predict the location and severity of damage for all potential *** achieve the most generalized surrogate model,the study explores different types of layers and hyperparameters of the training algorithm and employs state-of-the-art techniques to avoid overfitting and to accelerate the training *** approach’s effectiveness,efficiency,and applicability are demonstrated by two numerical *** study also verifies the robustness of the proposed approach on data sets with sparse and noisy measured ***,the proposed approach is a promising alternative to traditional approaches that rely on FE model updating and optimization algorithms,which can be computationally *** approach also shows potential for broader applications in structural damage detection.

关键词： vibration-based damage detection deep neural network full-scale structures finite element model updating noisy incomplete modal data

EP Aquarii:A New Picture of the Circumstellar Envelope

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Research in Astronomy and Astrophysics 2024年第3期24卷 83-105页

作者： Pham T.Nhung Do T.Hoai Pierre Darriulat Pham Tuan-anh Pham N.Diep nguyen B.ngoc Tran T.Thai Department of Astrophysics Vietnam National Space CenterVietnam Academy of Science and Technology18Hoang Quoc VietNghia DoCau GiayHa Noivietnam

New analyses of earlier ALMA observations of oxygen-rich AGB star EP Aquarii are presented,which complete a previously published analysis and offer a different interpretation of the morpho-kinematics of the circumstellar *** birth of the equatorial density enhancement(EDE)is shown to occur very close to the star where evidence for rotation has been *** to the star and where outflows have been observed:their interaction with the gas of the nascent EDE is seen to play an important role in the development of the wind and the evolution of its radial velocity from 8 to 10 km s^(-1) on the polar symmetry axis to~2 km s^(-1) at the *** implies complex morpho-kinematics:making reliable interpretations with reasonable confidence is *** particular,it questions an earlier interpretation implying the presence of a white dwarf companion orbiting the star at an angular distance of -0″4 from its *** proposes instead an interpretation in terms of a standard mass ejection associated with a shock wave leaving a void of emission in its *** Doppler velocity wings are seen to consist of two components,the upper velocity end of the global wind,reaching above±12 km s^(-1),and an effective line broadening,confined within 200 mas from the center of the star,reaching above±20 km s^(-1) and interpreted as caused by the pattern of shock waves resulting from the interaction between stellar pulsation and convective cell partition.

关键词： stars AGB and post-AGB-(stars) circumstellar matter-radio lines stars

Building 3D CityGML models of mining industrial structures using integrated UAV and TLS point clouds

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

International Journal of Coal Science & Technology 2023年第5期10卷 158-177页

作者： Canh Le Van Cuong Xuan Cao anh ngoc nguyen Chung Van Pham Long Quoc nguyen Department of Mine Surveying Hanoi University of Mining and GeologyHanoiVietnam Viettel Group Viettel Cyberspace CenterHanoiVietnam

Mining industrial areas with anthropogenic engineering structures are one of the most distinctive features of the real world.3D models of the real world have been increasingly popular with numerous applications,such as digital twins and smart factory *** this study,3D models of mining engineering structures were built based on the CityGML *** collecting spatial data,the two most popular geospatial technologies,namely UAV-SfM and TLS were *** accuracy of the UAV survey was at the centimeter level,and it satisfied the absolute positional accuracy requirement of creat-ing all levels of detail(LoD)according to the CityGML ***,the UAV-SfM point cloud dataset was used to build LoD 2 *** addition,the comparison between the UAV-SfM and TLS sub-clouds of facades and roofs indicates that the UAV-SfM and TLS point clouds of these objects are highly consistent,therefore,point clouds with a higher level of detail and accuracy provided by the integration of UAV-SfM and TLS were used to build LoD 3 *** resulting 3D CityGML models include 39 buildings at LoD 2,and two mine shafts with hoistrooms,headframes,and sheave wheels at LoD3.

关键词： 3D modelling CityGML-Mining industry UAV Terrestrial laser scanning Point cloud

Rapid and efficient characterization of cervical collagen orientation using linearly polarized colposcopic images

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Innovative Optical Health Sciences 2023年第5期16卷 21-31页

作者： Cat Phan ngoc Khuong Dung nguyen Huu Quang Hai Pham Thanh Long nguyen Quynh nguyen ngoc Duc Le Huynh Tien Tran Van Tu Ly anh Department of Applied Physics Ho Chi Minh City University of Technology 268 Ly Thuong Kiet Street District 10 Ho Chi Minh City 700000Vietnam Quality Management Department Tu Du Hospital 284 Cong QuynhDistrict 1 Ho Chi Minh City 700000Vietnam Department of Biomedical Engineering nguyen Tat Thanh University 300A Nguyen Tat Thanh District 4Ho Chi Minh City 700000Vietnam Faculty of Science and Technology Jean Monnet University 11 Rue du Docteur Annino Saint Etienne 42000France

Collagen provides tissue strength and structural *** fication of the orientated dispersion of collagen fibers is an important factor when studying the mechanical properties of the *** this study,for the first time,a new method for rapid characterization of the collagen fiber orientations of the cervix using linearly polarized light colposcopy is presented.A total of 24 colposcopic images were captured using a cross-polarized imaging system with white LED light *** the preprocessing stage,the Red channel of the RGB image was chosen,which contains no information of the blood vessels because of the low-absorption of blood cells in the red ***,which is an ImageJ plug-in,was used to estimate the local orientation of the collagen *** result shows that in the nonpregnant cervix,the middle zone(Zone 2)has circumferentially aligned collagen fibers while the inner zone(Zone 1)has randomly *** collagen fiber dispersion in Zone 2 is much smaller than that in Zone 1 at all four quadrants region(anterior,posterior,left,and right quadrant).This new analysis technique could potentially combine with diagnostic tools to provide a quantitative platform of collagen fibers in the clinic.

关键词： Cervix collagen fibers collagen orientation coherency cross-polarized imaging colposcopy

A Method of Integrating Length Constraints into Encoder-Decoder Transformer for Abstractive Text Summarization

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Intelligent Automation & Soft Computing 2023年第10期38卷 1-18页

作者： ngoc-Khuong nguyen Dac-Nhuong Le Viet-Ha nguyen anh-Cuong Le Faculty of Information Technology Haiphong UniversityHaiphong180000Vietnam Faculty of Information Technology VNU University of Engineering and TechnologyHanoi10000Vietnam Natural Language Processing and Knowledge Discovery Laboratory Faculty of Information TechnologyTon Duc Thang UniversityHo Chi Minh CityVietnam

Text summarization aims to generate a concise version of the original *** longer the summary text is,themore detailed it will be fromthe original text,and this depends on the intended ***,the problem of generating summary texts with desired lengths is a vital task to put the research into *** solve this problem,in this paper,we propose a new method to integrate the desired length of the summarized text into the encoder-decoder model for the abstractive text summarization *** length parameter is integrated into the encoding phase at each self-attention step and the decoding process by preserving the remaining length for calculating headattention in the generation process and using it as length embeddings added to theword *** conducted experiments for the proposed model on the two data sets,Cable News Network(CNN)Daily and NEWSROOM,with different desired output *** obtained results show the proposed model’s effectiveness compared with related studies.

关键词： Length controllable abstractive text summarization length embedding

The effectiveness of light emitting diode(LED)lamps in the offshore purse seine fishery in Vietnam

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Aquaculture and Fisheries 2023年第5期8卷 551-557页

作者： nguyen Dang Nhat Do Thanh Tien Truong Van Dan nguyen Duy Quynh tram nguyen Quang Lich Ho Dang Phuc nguyen ngoc Phuoc School of Engineering and Technology Hue University01 Dien Bien PhuHue City52000Viet Nam Faculty of Engineering and Food Technology University of Agriculture and ForestryHue University102 Phung HungHue City52000Viet Nam Faculty of Fisheries University of Agriculture and ForestryHue University102 Phung HungHue City52000Viet Nam Institute of Mathematics Vietnam Academy of Science and Technology18 Hoang Quoc VietCau Giay DistrictHanoi11300Viet Nam

Fishing with artificial light has become one of the most advanced,efficient,and common methods for the night-time purse seining in *** study evaluated the radiation spectrum,CIE chromaticity coordinates,correlated color temperature(CCT),catch rate,fuel consumption,and CO_(2)emissions when using Light emitting diode(LED)lamps(0.196 kW)in comparison with the use of metal halide(MH)lights(1 kW)in the offshore purse seine fishery in Quang Tri province,*** fishing efficiency of the purse seine fishing boats using LED lamps has increased 1.58 times in catch rate than MH lights,although the energy consumption of LED lamp is 4 times *** consumption of boats per trip using LED lamps was one third of that using MH *** use of LED reduced the radiation spectrum,especially the intense UV radiation which negatively affects the health of *** study also showed the potential of CO_(2)emission reduction up to 1.09 tons of CO_(2)per trip per boat from the use of LED lamps in the offshore purse seine fishing boats.

关键词： Light emitting diode(LED)lamp Metal halide(MH)light Offshore purse seine fishery Quang tri province Vietnam

A Novel Incremental Attribute Reduction Algorithm Based on Intuitionistic Fuzzy Partition Distance

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Systems Science & Engineering 2023年第12期47卷 2971-2988页

作者： Pham Viet anh nguyen ngoc Thuy nguyen Long Giang Pham Dinh Khanh nguyen The Thuy Graduate University of Science and Technology Vietnam Academy of Science and TechnologyHanoi100000Vietnam Institute of Information Technology Vietnam Academy of Science and TechnologyHanoi100000Vietnam HaUI Institute of Technology Hanoi University of IndustryHanoi100000Vietnam Faculty of Information Technology University of SciencesHue UniversityHue530000Vietnam AI Research Department Neurond Technology JSCHanoi100000Vietnam Information and Communication Technology Center Department of Information and CommunicationsBacninh790000Vietnam

Attribute reduction,also known as feature selection,for decision information systems is one of the most pivotal issues in machine learning and data *** based on the rough set theory and some extensions were proved to be efficient for dealing with the problemof attribute ***,the intuitionistic fuzzy sets based methods have not received much interest,while these methods are well-known as a very powerful approach to noisy decision tables,i.e.,data tables with the low initial classification ***,this paper provides a novel incremental attribute reductionmethod to dealmore effectivelywith noisy decision tables,especially for highdimensional *** particular,we define a new reduct and then design an original attribute reduction method based on the distance measure between two intuitionistic fuzzy *** should be noted that the intuitionistic fuzzypartitiondistance iswell-knownas aneffectivemeasure todetermine important *** interestingly,an incremental formula is also developed to quickly compute the intuitionistic fuzzy partition distance in case when the decision table increases in the number of *** formula is then applied to construct an incremental attribute reduction algorithm for handling such dynamic ***,some experiments are conducted on real datasets to show that our method is far superior to the fuzzy rough set based methods in terms of the size of reduct and the classification accuracy.

关键词： Incremental attribute reduction intuitionistic fuzzy sets partition distance measure dynamic decision tables

中越小学生对校园欺凌行为认识的比较以及教育对于提高越南小学生关于校园欺凌行为认识的作用

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

中越小学生对校园欺凌行为认识的比较以及教育对于提高越南小学生...

作者：阮玉英（nguyen ngoc anh tram）浙江师范大学

学位级别：硕士

校园欺凌是世界各国普遍存在且日益严重的问题。因此,研究者从心理学、教育学、社会学等不同角度选择了校园欺凌这个话题。目前在每个阶段的教育都存在着非常严重的校园欺凌现象,特别是在小学阶段。根据联合国儿童基金会(UNICEF,2018)... 详细信息

校园欺凌是世界各国普遍存在且日益严重的问题。因此,研究者从心理学、教育学、社会学等不同角度选择了校园欺凌这个话题。目前在每个阶段的教育都存在着非常严重的校园欺凌现象,特别是在小学阶段。根据联合国儿童基金会(UNICEF,2018)的一份报告:全球有一半的年轻人遭受过校园欺凌、打架和欺凌,这已经扰乱了全球1.5亿13-15岁儿童的教育。因此,全世界约有1.5亿学生报告说在学校和周边地区遭受同龄人的欺凌。事实上,这个数字并没有停止,校园欺凌已经成为全球教育的普遍问题。校园欺凌也严重影响了欺凌者和受欺凌者的健康和福祉。因此,探讨校园欺凌对于预防和减少校园欺凌的发生具有重要的现实意义。这项研究将致力于对中越小学生对校园欺凌行为的认识水平进行比较,并以越南小学生为例探讨教育对于提高小学生关于校园欺凌行为认识的作用。该研究采用问卷法考察中-越小学生对校园欺凌行为的认识,具体从以下五个方面设计问卷进行调查:(1)对校园欺凌行为概念的认识;(2)校园欺凌行为的成因;(3)校园欺凌行为的危害;(4)校园欺凌行为的反应形式;(5)预防校园欺凌行为的必要性及具体措施。该问卷共有85个题目,采用5点计分法计分,分数越高说明小学生对校园欺凌行为的认识越了解。通过对240名中-越小学生进行问卷调查发现,中、越小学生对校园欺凌行为的认识总体处于一般认知水平(中国,M=3.39;越南,M=3.28;均小于参考值3.40)。其中,对造成校园欺凌行为原因的认识(M=3.46)以及对预防校园欺凌行为必要性的认识(M=3.42)表现相对较好(均大于3.40)。另外,中国和越南小学生对校园欺凌行为的认识在性别、校际、成绩和操行上总体都不存在显著差异;在对中国和越南小学生的校园欺凌各个维度上的认识以及总体认识情况的比较中发现,只有在对校园欺凌行为的反应类型的认识上存在显著差异,表现为中国的小学生对校园欺凌行为的反应类型的认识更深,其它均未发现显著差异。针对研究一得出的小学生对校园欺凌行为的认识总体水平不高的事实,研究者以越南小学生为研究对象,探讨教育对提高小学生校园欺凌认识的作用,实验结果表明,有针对性的教育确实能够提高小学生对校园欺凌行为的认识。该研究还根据实验结果给学校、班主任、学校心理学家、家长和学生等提出教育建议。本研究为了解中越小学生对校园欺凌行为的认识做出了一定贡献,并通过对欺凌者与受欺凌者的教育进行研究,也取得了一些成果,这有助于学校在未来能够为学生提供更好的、健康的教学环境。

关键词：小学生校园欺凌对校园欺凌的认知中越比较

An improved CRNN for Vietnamese Identity Card Information Recognition

同方学位论文库评论

在线全文

同方学位论文库

学校读者我要写书评

暂无评论

Computer Systems Science & Engineering 2022年第2期40卷 539-555页

作者： Trinh Tan Dat Le Tran anh Dang nguyen Nhat Truong Pham Cung Le Thien Vu Vu ngoc Thanh Sang Pham Thi Vuong Pham The Bao Information Science Faculty Sai Gon UniversityHCM City700000Vietnam Faculty of Electrical&Electronics Engineering University of TechnologyHCM City700000Vietnam

This paper proposes an enhancement of an automatic text recognition system for extracting information from the front side of the Vietnamese citizen identity(CID)***,we apply Mask-RCNN to segment and align the CID card from the ***,we present two approaches to detect the CID card’s text lines using traditional image processing techniques compared to the EAST ***,we introduce a new end-to-end Convolutional Recurrent Neural Network(CRNN)model based on a combination of Connectionist Temporal Classification(CTC)and attention mechanism for Vietnamese text recognition by jointly train the CTC and attention objective functions *** length of the CTC’s output label sequence is applied to the attention-based decoder prediction to make the final label *** process helps to decrease irregular alignments and speed up the label sequence estimation during training and inference,instead of only relying on a data-driven attention-based encoder-decoder to estimate the label sequence in long *** may directly learn the proposed model from a sequence of words without detailed *** evaluate the proposed system using a real collected Vietnamese CID card dataset and find that our method provides a 4.28%in WER and outperforms the common techniques.

关键词： Vietnamese text recognition OCR CRNN BLSTM attention mechanism joint CTC-Attention