检索结果-南通市图书馆

Multi-Scale Feature Fusion Model for Bridge Appearance Defect Detection

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Big Data Mining and Analytics 2024年第1期7卷 1-11页

作者： Rong Pang Yan Yang Aiguo Huang Yan Liu Peng Zhang Guangwu Tang School of co.puting and Artificial Intelligence Southwest Jiaotong UniversityChengdu 611756China China Merchants Chongqing Road Engineering Inspection Center co. Ltd.Chongqing 400067China State Key Laboratory of Bridge Engineering Structural Dynamics Chongqing 400067China

Although the Faster Region-based co.volutional Neural Network(Faster R-CNN)model has obvious advantages in defect reco.nition,it still cannot overco.e challenging problems,such as time-co.suming,small targets,irregular shapes,and strong noise interference in bridge defect *** deal with these issues,this paper proposes a novel Multi-scale Feature Fusion(MFF)model for bridge appearance disease ***,the Faster R-CNN model adopts Region Of Interest(ROl)pooling,which omits the edge information of the target area,resulting in some missed detections and inaccuracies in both detecting and localizing bridge ***,this paper proposes an MFF based on regional feature Aggregation(MFF-A),which reduces the missed detection rate of bridge defect detection and improves the positioning accuracy of the target ***,the Faster R-CNN model is insensitive to small targets,irregular shapes,and strong noises in bridge defect detection,which results in a long training time and low reco.nition ***,a novel Lightweight MFF(namely MFF-L)model for bridge appearance defect detection using a lightweight network EfficientNetV2 and a feature pyramid network is proposed,which fuses multi-scale features to shorten the training speed and improve reco.nition ***,the effectiveness of the proposed method is evaluated on the bridge disease dataset and public co.putational fluid dynamic dataset.

关键词： defect detection Multi-scale Feature Fusion(MFF) Region Of Interest(ROl)alignment lightweight network

Train post‑derailment behaviours and co.tainment methods:a review

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Railway Engineering Science 2024年第1期32卷 59-80页

作者： Zhao Tang Yuwei Hu Shuangbu Wang Liang Ling Jianjun Zhang Kaiyun Wang State Key Laboratory of Traction Power Southwest Jiaotong UniversityChengdu 610031China Institute of Smart City and Intelligent Transportation Southwest Jiaotong UniversityChengdu 610031China National Centre for computer Animation Bournemouth UniversityDorsetUK

Railway accidents,particularly serious derailments,can lead to catastrophic ***,it is essential to prevent derailment escalation to reduce the likelihood of severe *** post-derailment behaviours and co.tainment methods play a critical role in preventing derailment escalation and providing passive safety protection and accident prevention in the event of a ***,despite the increasing attention on this field from academia and industry in recent years,there is a lack of systematic exploration and summarization of emerging applications and co.tainment methods in train post-derailment *** this reason,this paper presents a co.prehensive review of existing studies on train post-derailment behaviours,enco.passing various topics such as post-derailment co.tact-impact models,dynamic modelling and simulation techniques,and the primary factors influencing post-derailment ***,this review introduces and elucidates substitute guidance mechanisms(SGMs),which serve as railway-specific passive safety protection and accident prevention *** various types of SGMs are depicted,and their ongoing developments and applications are explored in *** review additionally points out several unresolved challenges including the adverse effects of SGMs,and proposes future research directions to advance the theoretical understanding and practical application of train post-derailment behaviours and co.tainment *** review seeks to be a valuable reference for railway industry professionals in preventing catastrophic derailment co.sequences through post-derailment co.tainment methods.

关键词： Railway passive safety Railway accident prevention Post-derailment behaviours Substitute guidance mechanisms(SGMs) co.tact-impact model

维普期刊数据库

Impact of sedimentary environmental evolution on organic matter enrichment in Upper Paleozoic marine–co.tinental transitional shales in the Shenfu Block， eastern Ordos Basin， China: evidence from geochemistry

在线全文

学校读者我要写书评

暂无评论

Journal of Earth Science 2024年

作者： Zhang Yanchao Yin Xiangdong Jiang Shu Ma Litao Li Jiaming School of Geoscience and Technology， southwest Petroleum University School of Sustainable Energy， China University of Geosciences Wuhan EnerTech-Drilling & Production co.，CNOOC Ltd

The shale gas exploration of marine-co.tinental transitional facies has gained great attention in China， which is viewed as another breakthrough following the marine shale gas prior to co.tinental shale gas. The Upper Paleozoic marine-co.tinental transitional shales have high TOC abundance with large total thickness in the eastern Ordos Basin， progressively beco.ing a favorable area for exploration of transitional shale gas. Nevertheless， co.pared with the relatively homogeneous marine shales， the marine-co.tinental transitional shales demonstrated strong anisotropy in the aspects of thickness， lithology and organic matter (OM). Thus， in this paper， based on the trace element and rock-eval data， we investigated the sedimentary facies， provenance， paleoclimate， paleobathymetry， and redox co.dition， as well as their co.trols on the enrichment of organic matter in the Shenfu Block， eastern Ordos basin. The results of trace element experiments indicate that the paleoenvironment of the Ordos Basin underwent a transition from a marine to a co.tinental environment， evolving from the Benxi Formation of the Carboniferous system to the Upper Shihezi Formation of the Permian system. The provenance area is characterised by the presence of high iron and high magnesium feldspar， with a medium degree of weathering. The paleoclimate experienced alternating periods of dry heat， relatively warm wet co.ditions， and dry heat. The paleowater depth decreased along the sedimentary profile. With regard to the REDOX environment index， the Taiyuan Formation is the stratum exhibiting the highest degree of reduction. The detection criteria for biomarkers were found to be co.sistent with the inorganic geochemical characteristics. The relationship between various trace elements and TOC indicates that redox co.ditions are the primary factors influencing the accumulation of organic matter in transitional shale. Organic matter is more abundant in warm and humid marine and terrestrial environments t

关键词：

Online review driven route planning of green co.d chain logistics: co.sidering service attribute preference

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Traffic and Transportation Engineering(English Edition) 2024年

作者： Qubo Yu Xuze Ye Xueying Jiang Shaoquan Ni Kun Liu Xiaowei Liu School of Transportation and Logistics， southwest Jiaotong University New co.ridor Logistics Group co.， Ltd.of Sichuan Port and Shipping Investment National Engineering Laboratory of Integrated Transportation Big Data Application Technology， southwest Jiaotong University

For the long-distance co.d chain transportation， this study focuses on green route planning in the co.d chain multimodal transport network， co.sidering the customers’ preferences on transport service attributes. To extract customers' preferences from online reviews， we build the bidirectional gated recurrent units (biGRU) and Logit framework. The co.fficients representing the preferences for transport service attributes are calculated and integrated into the route planning model. To enhance the reliability of the routes， we utilize fuzzy set theories to describe travel time， transport co.t co.fficient， and capacity as trapezoidal fuzzy numbers. Then， we develop a fuzzy route planning model aims to minimize transportation co.t， damage co.t， refrigeration co.t， carbon emission co.t， and reliability co.t. The numerical experiments are co.ducted on the new land and sea co.ridor in southwest China. The proposed method can plan the route which meet the customers’ preferences on transportation co.t， travel time， reliability and carbon emission. Moreover， logistics co.panies can strike a balance between co.t and reliability by setting co.fidence co.fficients for travel time and capacity. Furthermore， carbon trading policy implements the significant effect on carbon emission reduction in co.d chain logistics， when the trading price is 2 RMB per unit.

关键词：

Experimental and numerical simulation study on the erosion behavior of the elbow of gathering pipeline in shale gas field

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Petroleum Science 2024年第2期21卷 1257-1274页

作者： En-Bin Liu Shen Huang Ding-Chao Tian Lai-Min Shi Shan-Bi Peng He Zheng Petroleum Engineering School Southwest Petroleum UniversityChengdu 610500SichuanChina Chongqing Natural Gas Sales Center Natural Gas BranchSinopecChongqing400023China Zhejiang Zheneng Natural Gas Operation co. Ltd.Hangzhou 310000ZhejiangChina School of Civil Engineering and Architecture Southwest Petroleum UniversityChengdu 610500SichuanChina Louisiana Tech Univ Trenchless Technol Ctr599 Dan Reneau DrRustonLA71270USA southwest Oil and Gas Field Branch Natural Gas Research Institute Chengdu 610213SichuanChina

During the production period of shale gas, proppant particles and rock debris are produced together,which will seriously erode the elbows of gathering pipelines. In response to this problem, this paper takes the elbow of the gathering pipeline in the Changning Shale Gas Field as an example to test the erosion rate and material removal mechanism of the test piece at different angles of the elbow through experiments and co.pares the four erosion models with the experimental results. Through analysis, it is found that the best prediction model for quartz sand-carbon steel erosion is the Oka model. Based on the Oka model, FLUENT software was used to simulate and analyze the law of erosion of the elbow of the gas gathering pipeline under different gas flow velocities, gas gathering pressure, particle size, length of L1,and bending directions of the elbow. And a spiral pipeline structure is proposed to reduce the erosion rate of the elbow under the same working co.ditions. The results show that this structure can reduce erosion by 34%.

关键词： Shale gas Gas gathering pipeline Elbow Erosion CFD

维普期刊数据库

A Multi-AGV Routing Planning Method Based on Deep Reinforcement Learning and Recurrent Neural Network

在线全文

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2024年第7期11卷 1720-1722页

作者： Yishuai Lin Gang Hu Liang Wang Qingshan Li Jiawei Zhu the School of computer Science and Technology Xidian UniversityXi’an 710000China Suzhou Mingyi Intelligence Warehousing Information Technology co. Ltd.Kunshan 215300China the School of Information Engineering Chang’an UniversityXi’an 710000China

Dear Editor,This letter presents a multi-automated guided vehicles(AGV) routing planning method based on deep reinforcement learning(DRL)and recurrent neural network(RNN), specifically utilizing proximal policy optimi... 详细信息

关键词： network AGV Deep

维普期刊数据库

A co.munication Theory Perspective on Prompting Engineering Methods for Large Language Models

在线全文

学校读者我要写书评

暂无评论

Journal of computer Science & Technology 2024年第4期39卷 984-1004页

作者： Yuan-Feng Song Yuan-Qin He Xue-Fang Zhao Han-Lin Gu Di Jiang Hai-Jun Yang Li-Xin Fan AI Group WeBank Co.LtdShenzhen 518000China

The springing up of large language models(LLMs)has shifted the co.munity from single-task-orientated natural language processing(NLP)research to a holistic end-to-end multi-task learning *** this line of research endeavors in the area,LLM-based prompting methods have attracted much attention,partially due to the technological advantages brought by prompt engineering(PE)as well as the underlying NLP principles disclosed by various prompting *** supervised learning usually requires training a model based on labeled data and then making *** co.trast,PE methods directly use the powerful capabilities of existing LLMs(e.g.,GPT-3 and GPT-4)via co.posing appropriate prompts,especially under few-shot or zero-shot *** the abundance of studies related to the prompting and the ever-evolving nature of this field,this article aims to 1)illustrate a novel perspective to review existing PE methods within the well-established co.munication theory framework,2)facilitate a better/deeper understanding of developing trends of existing PE methods used in three typical tasks,and 3)shed light on promising research directions for future PE methods.

关键词： prompting method large language model co.munication theory

Microstructure and water-swelling mechanism of red-bed mudstone in the Xining region,Northeastern Tibetan Plateau

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 Rock Mechanics and Geotechnical Engineering 2024年第7期16卷 2537-2551页

作者： Huan Guan Yong Ren Sixiang Ling Xiyong Wu Tao Yu Xian Li Sen Wang Chunwei Sun School of Civil Engineering Southwest Jiaotong UniversityChengdu610031China Faculty of Geosciences and Engineering Southwest Jiaotong UniversityChengdu611756China Moe Key Laboratory of High-speed Railway Engineering Southwest Jiaotong UniversityChengdu610031China Guangzhou Metro Design and Research Institute co.Ltd. Guangzhou510010China Chengdu City co.struction Investment Group co.Ltd. Chengdu610036China

This paper examines the effect of the microstructure and electrical co.ductivity(EC)on the swelling ratio and pressure in red-bed mudstone sampled from arid areas in the Xining region in the northeastern Tibetan Plateau.A series of laboratory tests,including swelling experiments,X-ray diffraction(XRD),and scanning electron microsco.e(SEM),was carried out for mechanical and microstructural *** co.pled influence of the EC and microstructural parameters on the expansion ratio and pressure was investigated,and the weight co.fficients were discussed by the entropy weight *** results revealed an increasing exponential trend in EC,and the maximum swelling speed occurred at an EC of approximately 10 μS/*** addition,a method for predicting the expansion potential is proposed based on the microstructure,and its reliability is verified by co.paring with swelling experimental *** addition,acco.ding to the image analysis results,the ranges of the change in the clay minerals co.tent(CMC),the fractal dimension(FD),the average diameter(AD)of pores,and the plane porosity(PP)are 23.75%-53%,1.08-1.17,7.53-22.45 mm,and 0.62%-1.25%,***,mudstone swelling is negatively co.related with the plane porosity,fractal dimension and average diameter and is linearly co.related with the clay mineral ***,the weight values prove that the microstructural characteristics,including FD,AD,and PP,are the main factors influencing the expansion properties of red-bed mudstones in the Xining *** on the co.bination of macro and micro-analyses,a quantitative analysis of the swelling process of mudstones can provide a better reference for understanding the mechanism of expansion behavior.

关键词： Mudstones Microstructure Swelling characteristics Electrical co.ductivity(EC) Scanning electron microsco.e Water-swelling

Evolutionary Multi-Tasking Optimization for High-Efficiency Time Series Data Clustering

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2024年第2期29卷 343-355页

作者： Rui Wang Wenhua Li Kaili Shen Tao Zhang Xiangke Liao Xiangjiang Laboratory Changsha 410205China co.lege of Systems Engineering National University of Defense TechnologyChangsha 410073China Ant Group co. Ltd.Hangzhou 310000China co.lege of computer Science and Technology NUDTChangsha 410073China

Time series clustering is a challenging problem due to the large-volume,high-dimensional,and warping characteristics of time series *** clustering methods often use a single criterion or distance measure,which may not capture all the features of the *** paper proposes a novel method for time series clustering based on evolutionary multi-tasking optimization,termed i-MFEA,which uses an improved multifactorial evolutionary algorithm to optimize multiple clustering tasks simultaneously,each with a different validity index or distance ***,i-MFEA can produce diverse and robust clustering solutions that satisfy various preferences of *** on two artificial datasets show that i-MFEA outperforms single-objective evolutionary algorithms and traditional clustering methods in terms of co.vergence speed and clustering *** paper also discusses how i-MFEA can address two long-standing issues in time series clustering:the choice of appropriate similarity measure and the number of clusters.

关键词： time series clustering evolutionary multi-tasking multifactorial optimization clustering validity index distance measure