检索结果-南通市图书馆

Nuclear Science and Techniques 2024年第8期35卷 101-113页

作者： Hui‑Yin Shen Jing‑Long Zhang Jie Zhang Jian‑Hang Zhou college of Nuclear Technology and automation engineering Chengdu University of TechnologyChengdu 610059China Southwestern Institute of Physics P.O.Box 432Chengdu 610041China

Fast neutron flux measurements with high count rates and high time resolution have important applications in equipment such as *** this study,real-time neutron and gamma discrimination was implemented on a self-developed 500-Msps,12-bit digitizer,and the neutron and gamma spectra were calculated directly on an FPGA.A fast neutron flux measurement system with BC-501A and EJ-309 liquid scintillator detectors was developed and a fast neutron measurement experiment was successfully performed on the HL-2 M tokamak at the Southwestern Institute of Physics,*** experimental results demonstrated that the system obtained the neutron and gamma spectra with a time accuracy of 1 *** count rates of up to 1 Mcps,the figure of merit was greater than 1.05 for energies between 50 keV and 2.8 MeV.

关键词： Neutron-gamma discrimination Liquid scintillation detector Real-time spectrum analyzer

Learning to Branch in Combinatorial Optimization With Graph Pointer Networks

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1期11卷 157-169页

作者： Rui Wang Zhiming Zhou Kaiwen Li Tao Zhang Ling Wang Xin Xu Xiangke Liao IEEE the Xiangjiang Laboratory and the college of Systems engineering National University of Defense TechnologyChangsha 410073China the Institute of automation Chinese Academy of SciencesBeijing 100190China the college of Systems engineering National University of Defense TechnologyChangsha 410073 the Hunan Key Laboratory of Multi-Energy System Intelligent Interconnection Technology Changsha 410073China the Department of automation Tsinghua UniversityBeijing 100084China the college of Intelligence Science and Technology National University of Defense TechnologyChangsha 410073China the college of Computer Science and Technology National University of Defense TechnologyChangsha 410073China

Traditional expert-designed branching rules in branch-and-bound(B&B) are static, often failing to adapt to diverse and evolving problem instances. Crafting these rules is labor-intensive, and may not scale well with complex *** the frequent need to solve varied combinatorial optimization problems, leveraging statistical learning to auto-tune B&B algorithms for specific problem classes becomes attractive. This paper proposes a graph pointer network model to learn the branch rules. Graph features, global features and historical features are designated to represent the solver state. The graph neural network processes graph features, while the pointer mechanism assimilates the global and historical features to finally determine the variable on which to branch. The model is trained to imitate the expert strong branching rule by a tailored top-k Kullback-Leibler divergence loss function. Experiments on a series of benchmark problems demonstrate that the proposed approach significantly outperforms the widely used expert-designed branching rules. It also outperforms state-of-the-art machine-learning-based branch-and-bound methods in terms of solving speed and search tree size on all the test instances. In addition, the model can generalize to unseen instances and scale to larger instances.

关键词： Branch-and-bound(B&B) combinatorial optimization deep learning graph neural network imitation learning

Flow field characteristics and drag reduction performance of high-low velocity stripes on the biomimetic imbricated fish scale surfaces

在线全文

学校读者我要写书评

暂无评论

Biosurface and Biotribology 2024年第3期10卷 132-141页

作者： Dengke Chen XianXian Cui Huawei Chen college of Transportation Ludong UniversityYantaiShandongChina School of mechanical engineering and automation Beihang UniversityBeijingChina Advanced Innovation Center for Biomedical engineering Beihang UniversityBeijingChina

Improving energy efficiency and cost reduction is a perennial challenge in *** biological systems have evolved unique functional surfaces or special physiological functions over centuries to adapt to their complex *** these biological wonders,fish,one of the oldest vertebrate groups,has garnered significant attention due to its exceptional fluid dynamics *** are actively exploring the potential of fish skin's distinctive structural and material characteristics in reducing *** this study,models of biomimetic imbricated fish scale are established,and the evolution characteristics of the flow field and drag reduction performance on these bionic surfaces are *** results showed a close relationship between the high-low velocity stripes generated and the fluid motion by the imbricated fish scale *** stripes'prominence increases with the spacing of the adjacent scales and tilt angle of the fish scale,and the velocity amplitude of the stripes decreases as the exposed length of the imbricated fish scale surface ***,the biomimetic imbricated fish scale surface can decrease the velocity gradient and thereby reduce the wall shear *** insights gained from the fish skin-inspired imbricated fish surface provide valuable perspectives for an in-depth analysis of fish hydrodynamics and offer fresh inspiration for drag reduction and antifouling strategies in engineering applications.

关键词： bionic surface drag reduction

A Differential Privacy Federated Learning Scheme Based on Adaptive Gaussian Noise

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 Modeling in engineering & Sciences 2024年第2期138卷 1679-1694页

作者： Sanxiu Jiao Lecai Cai Xinjie Wang Kui Cheng Xiang Gao college of automation and Information engineering Sichuan University of Science and EngineeringYibin644000China Sanjiang Research Institute of Artificial Intelligence and Robotics Yibin UniversityYibin644000China college of mechanical engineering Sichuan University of Science and EngineeringYibin644000China

As a distributed machine learning method,federated learning(FL)has the advantage of naturally protecting data *** keeps data locally and trains local models through local data to protect the privacy of local *** federated learning method effectively solves the problem of artificial Smart data islands and privacy protection ***,existing research shows that attackersmay still steal user information by analyzing the parameters in the federated learning training process and the aggregation parameters on the server *** solve this problem,differential privacy(DP)techniques are widely used for privacy protection in federated ***,adding Gaussian noise perturbations to the data degrades the model learning *** address these issues,this paper proposes a differential privacy federated learning scheme based on adaptive Gaussian noise(DPFL-AGN).To protect the data privacy and security of the federated learning training process,adaptive Gaussian noise is specifically added in the training process to hide the real parameters uploaded by the *** addition,this paper proposes an adaptive noise reduction *** the convergence of the model,the Gaussian noise in the later stage of the federated learning training process is reduced *** paper conducts a series of simulation experiments on realMNIST and CIFAR-10 datasets,and the results show that the DPFL-AGN algorithmperforms better compared to the other algorithms.

关键词： Differential privacy federated learning deep learning data privacy

Drag reduction capacity of multi-scale and multi-level riblet in turbulent flow

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Biosurface and Biotribology 2024年第1期10卷 7-15页

作者： Dengke Chen Wenhao Li Yichen Zhao Jinhai Liu Xianxian Cui Zehui Zhao Xiaolin Liu Huawei Chen college of Transportation Ludong UniversityYantaiShandongChina School of mechanical engineering and automation Beihang UniversityBeijingChina Advanced Innovation Center for Biomedical engineering Beihang UniversityBeijingChina

For high-speed moving objects,drag reduction has been a prolonged major *** address this problem,passive and negative strategies have been proposed in the preceding *** integration of creatures and nature has been continuously perfected during biological *** structure characteristics,material properties,and special functions of marine organisms can provide inexhaustible inspirations to solve this intractable problem of drag ***,a simple and low-cost laser ablation method was proposed.A multi-scale and multi-level riblet(MSLR)surface inspired by the denticles of the sharkskin was fabricated by controlling the density of the laser path and ablation *** morphology and topographic features were characterised using an electron microscope and a scanning white-light interfering ***,the drag reduction capacity of the bionic riblet surface was measured in a circulating water ***,the mechanism of drag reduction was analysed by the computational fluid dynamics(CFD)*** results show that the MSLR surface has a stable drag reduction capacity with an increase in Reynold(Re)number which was contributed by high-low velocity stripes formed on the MSLR *** study can provide a reference for fabricating spatial riblets with efficient drag reduction at different values of Re and improving marine antifouling.

关键词： BIOMIMICS BIONIC STRUCTURE DRAG REDUCTION

A Two-Layer Encoding Learning Swarm Optimizer Based on Frequent Itemsets for Sparse Large-Scale Multi-Objective Optimization

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2024年第6期11卷 1342-1357页

作者： Sheng Qi Rui Wang Tao Zhang Xu Yang Ruiqing Sun Ling Wang college of Systems engineering National University of Defense TechnologyChangsha 410073China Xiangjiang Laboratory Changsha 410205China college of Computer National University of Defense TechnologyChangsha 410073China Department of automation Tsinghua UniversityBeijing 100084China IEEE

Traditional large-scale multi-objective optimization algorithms(LSMOEAs)encounter difficulties when dealing with sparse large-scale multi-objective optimization problems(SLM-OPs)where most decision variables are *** a result,many algorithms use a two-layer encoding approach to optimize binary variable Mask and real variable Dec ***,existing optimizers often focus on locating non-zero variable posi-tions to optimize the binary variables ***,approxi-mating the sparse distribution of real Pareto optimal solutions does not necessarily mean that the objective function is *** data mining,it is common to mine frequent itemsets appear-ing together in a dataset to reveal the correlation between *** by this,we propose a novel two-layer encoding learning swarm optimizer based on frequent itemsets(TELSO)to address these *** mined the frequent terms of multiple particles with better target values to find mask combinations that can obtain better objective values for fast ***-mental results on five real-world problems and eight benchmark sets demonstrate that TELSO outperforms existing state-of-the-art sparse large-scale multi-objective evolutionary algorithms(SLMOEAs)in terms of performance and convergence speed.

关键词： Evolutionary algorithms learning swarm optimiza-tion sparse large-scale optimization sparse large-scale multi-objec-tive problems two-layer encoding.

Monocular Distance Estimated Based on PTZ Camera

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 3417-3433页

作者： Qirui Zhong Xiaogang Cheng Yuxin Song Han Wang Bell Honors School Nanjing University of Posts and Telecommunications Nanjing210003China college of Telecommunications and Information engineering Nanjing University of Posts and TelecommunicationsNanjing210003China college of automation and college of Artificial Intelligence Nanjing University of Posts and TelecommunicationsNanjing210003China

This paper introduces an intelligent computational approach for extracting salient objects fromimages and estimatingtheir distance information with PTZ (Pan-Tilt-Zoom) cameras. PTZ cameras have found wide applications innumerous public places, serving various purposes such as public securitymanagement, natural disastermonitoring,and crisis alarms, particularly with the rapid development of Artificial Intelligence and global infrastructuralprojects. In this paper, we combine Gauss optical principles with the PTZ camera’s capabilities of horizontal andpitch rotation, as well as optical zoom, to estimate the distance of the *** present a novel monocular objectdistance estimation model based on the Focal Length-Target Pixel Size (FLTPS) relationship, achieving an accuracyrate of over 95% for objects within a 5 km range. The salient object extraction is achieved through a simplifiedconvolution kernel and the utilization of the object’s RGB features, which offer significantly faster computingspeeds compared to Convolutional Neural Networks (CNNs). Additionally, we introduce the dark channel beforethe fog removal algorithm, resulting in a 20 dB increase in image definition, which significantly benefits distanceestimation. Our system offers the advantages of stability and low device load, making it an asset for public securityaffairs and providing a reference point for future developments in surveillance hardware.

关键词： Public security location information PTZ camera image processing

Electrical tree degradation of MgO/epoxy resin composites at different voltage frequencies

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

High Voltage 2024年第3期9卷 581-590页

作者： Yunxiao Zhang Wenxin Lin Yuanxiang Zhou Weiwei Xing Jiayu Cheng Chenyuan Teng college of Electrical engineering and automation Fuzhou UniversityFujian Key Laboratory of New Energy Generation and Power ConversionFuzhouChina State Key Laboratory of Control and Simulation of Power Systems and Generation Equipment Department of Electrical EngineeringTsinghua UniversityBeijingChina

Electrical tree degradation is one of the main causes of insulation failure in high-frequency *** tree degradation is studied on pure epoxy resin(EP)and MgO/EP composites at frequencies ranging from 50 Hz to 130 *** results show that the tree initiation voltage of EP decreases,while the growth rate and the expansion coefficient increase with ***,the bubble phenomenon at high frequencies in EP composites is *** with trap distribution character-istics within the material,the intrinsic mechanism of epoxy composites to inhibit the growth of the electrical tree at different frequencies is *** can be concluded that more deep traps and blocking effect are introduced by doping nano-MgO into EP bulks,which can improve the electrical tree resistance performance of EP composites in a wide frequency range.

关键词： composites tree Electrical

Concept, configuration, and application of on-board nanogrids for use in electric vehicles

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Technological Sciences) 2024年第7期67卷 2053-2073页

作者： ZHAO Bing LI Yong WU Fang WAN JiangHu CAI JiuQing YANG HaiTao LIU Fang college of Electrical and Information engineering Hunan UniversityChangsha 410082China Wuhan Second Ship Design and Research Institute Wuhan 430000China School of automation Central South UniversityChangsha 410083China

Against the backdrop of increasingly serious climate change, researchers are attempting to extend macroscale carbon reduction research to smaller scales. With the rapid development and widespread application of electric vehicles(EVs), an effective approach for carbon reduction based on EVs has the potential to be developed. To coordinate and manage the EV platform in diverse application scenarios, the concept of an on-board nanogrid(OBNG), in which a nanogrid is combined with the EV, is proposed and defined, and the characteristics summarized. A configuration that includes a physical layer with the on-board hardware system;an information layer for logical control, energy management, and communication coordination;and an application layer that can cope with different working environments is proposed. A detailed introduction to the basic architecture and management mode of each layer is provided along with information concerning the relevant technologies for coordinated operation. New ideas and approaches to improve the existing performance are proposed, and finally, combined with a background of smart and low-carbon cities, major application scenarios are envisioned.

关键词： electric vehicle nanogrid energy management hybrid energy storage power electronics converter

Doped low-density parity-check codes

在线全文

学校读者我要写书评

暂无评论

Digital Communications and Networks 2024年第1期10卷 217-226页

作者： Yong Li Rui Liu Xianlong Jiao Youqiang Hu Zhen Luo Francis C.M.Lau college of Computer Science Chongqing UniversityChongqing400044China School of automation Chongqing UniversityChongqing400044China college of Electronic and Information engineering Southwest UniversityChongqing400715China Future Wireless Networks and IoT Focusing Area Department of Electronic and Information EngineeringHong Kong Polytechnic UniversityHong KongChina

In this paper,we propose a doping approach to lower the error floor of Low-Density Parity-Check(LDPC)*** doping component is a short block code in which the information bits are selected from the coded bits of the dominant trapping sets of the LDPC ***,an algorithm for selecting the information bits of the short code is proposed,and a specific two-stage decoding algorithm is *** results demonstrate that the proposed doped LDPC code achieves up to 2.0 dB gain compared with the original LDPC code at a frame error rate of 10^(-6)Furthermore,the proposed design can lower the error floor of original LDPC Codes.

关键词： LDPC codes Doped LDPC codes Tanner graph Quadratic residue codes Trapping sets