期刊文献+
任意字段
题名或关键词
题名
关键词
摘要
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息
共找到39,815篇文章
< 1 2 … 250 >
每页显示 20 50 100
已选0条
引用分析
参考文献
引证文献
 统计分析
检索结果
已选文献
显示方式：
An Intelligent Algorithm for Dynamic Scheduling of Parallel Machines Considering Multi-Task Collaboration in Order Processing 认领 引用
1
作者 Pei Xie Xiaoying Yang +2 位作者 Bo Li Zhijie Pei Fenghai Yang 《Computers, Materials & Continua》 SCIE EI 2026年第9期813-836,共24页
To address the critical requirements for collaborative delivery of multiple tasks within each order in personalized mass customization,this paper develops a dynamic parallel machine scheduling model that accounts for ... To address the critical requirements for collaborative delivery of multiple tasks within each order in personalized mass customization,this paper develops a dynamic parallel machine scheduling model that accounts for stochastic machine failures and order priorities,thereby more accurately reflecting the uncertainties and complexities of real-world production environments.A dual-objective optimization framework is adopted to minimize both the makespan(maximum task completion time)and the variance of task completion times,aiming to improve the coordination and reliability of intra-order task delivery.An adaptive weighted reward function is designed to balance overall scheduling efficiency with consistency among tasks during reinforcement learning training.To tackle the challenges posed by partially observable Markov decision processes(POMDP)induced by unexpected machine breakdowns,a Gated Recurrent Unit(GRU)-embedded Proximal Policy Optimization(PPO)intelligent scheduling algorithm is proposed.The algorithm incorporates an Action Masking mechanism to prevent invalid scheduling actions,while the GRU module captures historical state sequences to enhance perception of dynamic production environments.Extensive validation on benchmark datasets,along with comparisons against traditional heuristic algorithms,metaheuristic algorithms,and other deep reinforcement learning methods,demonstrates that the proposed approach achieves robust convergence,high resilience,and strong generalization across both static and dynamic scenarios,significantly improving coordinated delivery performance of order tasks.Overall,the proposed method not only provides an efficient and scalable real-time decision-making solution for Parallel Machine Scheduling Problems(PMSP)but also offers new theoretical and practical insights for optimizing complex production scheduling in intelligent manufacturing systems. 展开更多
关键词 Parallel machine scheduling problems dynamic scheduling gated recurrent unit proximal policy optimization coordinated delivery
暂未订购 下载PDF
Microseismic signal processing and rockburst disaster identification:A multi-task deep learning and machine learning approach 认领 引用 被引量：1
2
作者 Chunchi Ma Weihao Xu +3 位作者 Xuefeng Ran Tianbin Li Hang Zhang Dongwei Xing 《Journal of Rock Mechanics and Geotechnical Engineering》 SCIE CSCD 2026年第1期441-456,共16页
Underground engineering projects such as deep tunnel excavation often encounter rockburst disasters accompanied by numerous microseismic events.Rapid interpretation of microseismic signals is crucial for the timely id... Underground engineering projects such as deep tunnel excavation often encounter rockburst disasters accompanied by numerous microseismic events.Rapid interpretation of microseismic signals is crucial for the timely identification of rockbursts.However,conventional processing encompasses multi-step workflows,including classification,denoising,picking,locating,and computational analysis,coupled with manual intervention,which collectively compromise the reliability of early warnings.To address these challenges,this study innovatively proposes the“microseismic stethoscope"-a multi-task machine learning and deep learning model designed for the automated processing of massive microseismic signals.This model efficiently extracts three key parameters that are necessary for recognizing rockburst disasters:rupture location,microseismic energy,and moment magnitude.Specifically,the model extracts raw waveform features from three dedicated sub-networks:a classifier for source zone classification,and two regressors for microseismic energy and moment magnitude estimation.This model demonstrates superior efficiency compared to traditional processing and semi-automated processing,reducing per-event processing time from 0.71 s to 0.49 s to merely 0.036 s.It concurrently achieves 98%accuracy in source zone classification,with microseismic energy and moment magnitude estimation errors of 0.13 and 0.05,respectively.This model has been well applied and validated in the Daxiagu Tunnel case in Sichuan,China.The application results indicate that the model is as accurate as traditional methods in determining source parameters,and thus can be used to identify potential geomechanical processes of rockburst disasters.By enhancing the signal processing reliability of microseismic events,the proposed model in this study presents a significant advancement in the identification of rockburst disasters. 展开更多
关键词 Underground engineering Microseismic signal processing Deep learning Multi-task Rockburst identification
暂未订购 下载PDF
Parallel divergence with shared barriers,and niche separation between two sympatric avian species groups 认领 引用 被引量：1
3
作者 Lei Wu Huan Wang +10 位作者 Yanzhu Ji Ali Haghani Yan Hao Dezhi Zhang Gang Song Yalin Cheng Martin Päckert Jochen Martens Chenxi Jia Per Alström Fumin Lei 《Journal of Systematics and Evolution》 SCIE CSCD 2026年第1期77-94,共18页
Geographic barriers and geological historical events may play pivotal roles in driving allopatric divergence among closely related species.Here,we investigate the genomic divergence patterns and ecological niche separ... Geographic barriers and geological historical events may play pivotal roles in driving allopatric divergence among closely related species.Here,we investigate the genomic divergence patterns and ecological niche separation of the Willow Tit Poecile montanus and the Marsh Tit P.palustris species groups in China,and their ecological niche separation across East Asia.Through comprehensive genomic sequencing,population genomic analysis,and integration of public occurrence data,we unveil striking parallels in the geographic divergence patterns between these two species groups.Notably,both species exhibit multiple divergent lineages in China,with similar spatial distributions of geneflow barriers.Furthermore,our analysis reveals unique evolutionary histories in the southwestern clades of both species groups,highlighting the intricate interplay between historical distribution dynamics,ecological preferences,and genetic divergence.Our study significantly enhances our understanding of the processes underlying the diversification of closely related widespread species within the framework of shared geographical constraints,and stresses the need for a taxonomic revision. 展开更多
关键词 biogeography East Asia Marsh Tit parallel divergence Willow Tit
Multi-task U-net inversion of synthetic look-ahead logging-while-drilling data 认领 引用
4
作者 Shun Zhang Wen-Xiu Zhang +3 位作者 Wen-Xuan Chen Peng-Fei Liang Wen-Yang Wang Xing-Han Li 《Petroleum Science》 SCIE EI CAS CSCD 2026年第4期1908-1928,共21页
Electromagnetic look-ahead logging while drilling instruments detect the electrical characteristics of undrilled formations,enabling proactive decision-making.Real-time geological insight ahead of the drill bit is cri... Electromagnetic look-ahead logging while drilling instruments detect the electrical characteristics of undrilled formations,enabling proactive decision-making.Real-time geological insight ahead of the drill bit is critical for effective geosteering.This study introduces a multi-task U-net neural network that simultaneously inverts multiple formation parameters real-time.Six datasets,each corresponding to different electromagnetic components,were used to train six specialized neural networks.All networks exhibited rapid convergence and successfully inverted 60,000 sample in 15 s,satisfying real-time requirements.Residual and relative error analyses reveal that the multi-component network delivers the highest accuracy.Sensitivity analysis shows that coaxial and coplanar components are more sensitive to conductivity variations,whereas coaxial and cross-components excel at resolving interface positions.The yy component displays the strongest sensitivity to anisotropy.Compared with the traditional Levenberg-Marquardt algorithm,the proposed method demonstrates improved accuracy and efficiency.Moreover,the Levenberg-Marquardt inversion with the neural network output as initial models further enhances accuracy.Benchmark comparisons reveal that the multi-task U-net outperforms various mainstream machine learning and deep learning models,including LSTM,FCN,ResNet,and XGBoost,in both inversion accuracy and generalization.Moreover,sensitivity analyses to noise and near-bit geological complexity reveal that,while the proposed model experiences some performance degradation under high noise levels or highly heterogeneous backgrounds,it maintains strong robustness under moderate noise conditions and achieves reliable inversion results in two-layer geological settings.These results establish the multi-task U-net as a fast,accurate,and robust tool for real-time electromagnetic look-ahead inversion in geosteering applications. 展开更多
关键词 Multi-task U-net Look-ahead Anisotropy Multiple components Inversion
暂未订购 下载PDF
In-Sensor Reservoir Computing Employing Reconfigurable Optoelectronic Transistors for Multi-Task Learning 认领 引用
5
作者 Shanshan Jiang Hainan Zhang +4 位作者 Kesheng Wang Shuo Cheng Can Fu Huanhuan Wei Gang He 《Rare Metals》 SCIE EI CAS CSCD 2026年第4期629-639,共11页
The edge deployment of artificial intelligence has driven the exploitation of compact,energy-efficient information processing systems that integrate sensing,memory,and multi-task processing functions.However,conventio... The edge deployment of artificial intelligence has driven the exploitation of compact,energy-efficient information processing systems that integrate sensing,memory,and multi-task processing functions.However,conventional vision systems suffer from significant energyime overhead,extra hardware costs,and an unaffordable algorithm.Herein,we demonstrate an in-sensor computing system employing reconfigurable optoelectronic transistors(ROETs)for multi-task learning.These transistors exhibit reconfigurable volatile and nonvolatile characteristics under both optical and electrical stimuli.Capitalizing on this reconfigurability,we establish an in-sensor reservoir computing(RC)system operating in multi-signal modes:volatile dynamics function as the reservoir,whereas nonvolatile properties configure the readout layer.The abundant optoelectronic reservoir states display exceptional feature separability and prolonged stability in the ambient atmosphere.Such a reliable RC system successfully achieves multi-task processing of images.Notably,under the optoelectronic coordination mode,it effectively alleviates feature degradation while sustaining consistently high recognition accuracy.Furthermore,the system exhibits remarkable dynamic information processing capabilities,achieving recognition accuracies of 89.02%for dynamic gestures and 96.04%for moving vehicles recognition,respectively.Supplemental functionalities,including light adaptation and image sharpening,are also implemented.This work presents a configurable multimodal platform featuring a flexible in-sensor reservoir computing architecture,providing a potential solution for efficient multi-task processing. 展开更多
关键词 in-sensor reservoir computing multi-task learning neuromorphic applications optoelectronic transistors reconfigurable devices
暂未订购 下载PDF
FTCSEM—A FORTRAN-based parallelized 1D CSEM forward and inversion program for arbitrary source-receiver geometry 认领 引用
6
作者 Wei-ying Chen Si-xu Han +2 位作者 Wan-ting Song Yu-lian Zhu Zheng Liu 《Applied Geophysics》 SCIE CSCD 2026年第2期693-709,870,共17页
This study introduces FTCSEM,a FORTRAN-based,parallelized one-dimensional controlledsource electromagnetic(CSEM)forward modeling and inversion software capable of accommodating arbitrary source-receiver confi guration... This study introduces FTCSEM,a FORTRAN-based,parallelized one-dimensional controlledsource electromagnetic(CSEM)forward modeling and inversion software capable of accommodating arbitrary source-receiver confi gurations.In comparison to existing one-dimensional CSEM tools,FTCSEM incorporates several signifi cant enhancements:it supports transmitters of diverse shapes,quantities,and spatial locations;permits receivers to be positioned flexibly on the surface,subsurface,or in the atmosphere;facilitates simulations and inversions in both frequency and time domains;integrates an adaptive regularized inversion algorithm with multiple model constraints;and leverages GPU-accelerated parallel computing to attain high computational efficiency.Validation through numerical experiments and field data inversion confirms the program’s accuracy and practical applicability.The findings indicate that FTCSEM performs robustly in complex geoelectric environments,multi-source and multi-receiver arrangements,as well as multi-component joint inversion scenarios,thereby offering a versatile and powerful tool for advancing CSEM research and applications. 展开更多
关键词 CSEM forward modeling regularized inversion parallel computing program
暂未订购 下载PDF
Advancements and theoretical foundations of cable-driven parallel robots:A comprehensive overview 认领 引用
7
作者 Kamran Joyo Han Yuan +2 位作者 Jun Wu Wenfu Xu Honghao Yue 《Chinese Journal of Mechanical Engineering》 SCIE EI CAS CSCD 2026年第3期280-305,共26页
Parallel robotic mechanisms using cables instead of rigid limbs are termed cable-driven parallel robots(CDPRs).Electric motors and pulley mechanisms actuate cables to provide motion for an end-effector in a cable robo... Parallel robotic mechanisms using cables instead of rigid limbs are termed cable-driven parallel robots(CDPRs).Electric motors and pulley mechanisms actuate cables to provide motion for an end-effector in a cable robot.Consequently,CDPRs have emerged as indispensable tools across a spectrum of industrial and technological domains,including astronomy,aerospace,logistics,simulators,and rehabilitation.Their inherent compatibility with the evolving concept of rigid-flexible fusion places CDPRs at the forefront of cutting-edge robotics research.This comprehensive paper aims to consolidate the core theories and advancements underpinning CDPRs,en-compassing key aspects such as configuration design,cable-force distribution,workspace and stiffness analysis,performance evaluation,optimisation techniques,and motion control.We provide in-depth insights into kine-matic modelling,workspace exploration,and cable-force solutions.Furthermore,the paper delves into the in-tricacies of stiffness and dynamic modelling,presenting a range of analytical methods to elucidate their effects on CDPR performance.Addressing reliability concerns and developing a unified control framework are identified as essential in ensuring the practical deployment of CDPRs in real-world scenarios.This research paper offers a comprehensive overview of the theories and advancements in CDPRs,identifying critical areas for further re-search and development to unlock the full potential of these versatile and high-performance robotic systems. 展开更多
关键词 Cable-driven parallel robots Optimisation Path generation Statics Machine learning
暂未订购 下载PDF
Optimal design of a novel dual-axis parallel solar tracker 认领 引用
8
作者 Bin Zhu Liping Wang +2 位作者 Jun Wu Hengchun Cui Yanling Tian 《Chinese Journal of Mechanical Engineering》 SCIE EI CAS CSCD 2026年第2期167-179,共13页
Being renewable and readily available,solar energy has gained significant attention in addressing the global energy crisis and climate change.The efficiency of solar-energy harvesting using a concentrator depends on t... Being renewable and readily available,solar energy has gained significant attention in addressing the global energy crisis and climate change.The efficiency of solar-energy harvesting using a concentrator depends on the angle between the incident sunlight and the solar concentrator.Therefore,a solar-energy collection system equipped with a solar tracker that follows the apparent motion of the sun offers the highest collection efficiency.In this study,a novel solar tracker with a parallel mechanism is proposed based on the line graph method.The proposed parallel solar tracker(PST) features a main column with passive movements and two UPU chains that share a common constraint.This design enhances the rotational workspace,stiffness,and load-bearing capacity of the system.To solve the forward kinematics problem of the PST efficiently and accurately,a geometric elimination method is employed,converting the three-dimensional kinematics into a simpler planar problem.This allows the forward kinematics problem to be solved analytically using planar equations.By considering key performance indices,such as the effective workspace,transmission,and manipulability,the structural parameters of the PST are optimized in two steps,thereby identifying the optimal region in the design space.Finally,the computational efficiency and accuracy of both the forward and inverse kinematic solutions for the PST with optimized structural parameters are validated,demonstrating their potential for use in real-time control systems.The proposed novel solar tracker has high stiffness and load-bearing capacity.The study provides a solid foundation for improving the efficiency of solar energy utilization. 展开更多
关键词 Solar tracker Parallel mechanism Forward kinematics Effective workspace Transmissibility Manipulability
暂未订购 下载PDF
The three-dimensional meshfree numerical manifold method based on parallel computing 认领 引用
9
作者 Keqin Zhang Wei Wu +2 位作者 Danfeng Zhang Yanfei Kang Hehua Zhu 《Acta Mechanica Sinica》 SCIE EI CAS CSCD 2026年第2期360-371,共12页
The numerical manifold method,extensively utilized in numerical computations,faces significant challenges in generating complex manifold elements,particularly for three-dimensional applications.To overcome this challe... The numerical manifold method,extensively utilized in numerical computations,faces significant challenges in generating complex manifold elements,particularly for three-dimensional applications.To overcome this challenge,the meshfree numerical manifold method is developed by integrating the moving least-squares method into the numerical manifold method,effectively bypassing the need for meshing complex geometric objects.However,the implementation of the moving least-squares method introduces computational efficiency issues.To mitigate these,parallel computing methods have been incorporated,resulting in a tenfold increase in the speed of assembling the stiffness matrix with central processing unit parallelism,and a twentyfold increase with graphics processing unit parallelism.The static mechanical system equations for the meshfree numerical manifold method are derived using the Galerkin method.The method’s effectiveness and accuracy are then validated through a series of numerical experiments.The experiments demonstrated that the meshfree numerical manifold method achieves a high precision with minimal nodes and integration points.Additionally,positioning nodes outside the domain significantly improves computational accuracy at the boundaries. 展开更多
关键词 Meshfree numerical manifold method Three-dimensional computation Parallel computation Elastostatics Moving least-squares
暂未订购 下载PDF
A high-performance parallel algorithm based on problem independent machine learning(PIML)for large-scale topology optimization 认领 引用
10
作者 Xinyu Ma Mengcheng Huang +4 位作者 Zongliang Du Yilin Guo Chang Liu Yue Mei Xu Guo 《Acta Mechanica Sinica》 SCIE EI CAS CSCD 2026年第3期197-213,共17页
Although supplying extensive design space,the curse of dimensionality restricts the widespread application of largescale topology optimization in practical engineering.Various acceleration techniques have been integra... Although supplying extensive design space,the curse of dimensionality restricts the widespread application of largescale topology optimization in practical engineering.Various acceleration techniques have been integrated with topology optimization,achieving significant attention and progress in large-scale problems.This work aims to investigate how much benefit can be obtained by combining parallel computing and machine learning techniques to enhance the efficiency of large-scale topology optimization algorithms.Accordingly,a parallel problem independent machine learning(PIML)-enhanced topology optimization method is proposed.The PIML model substantially reduces the dimension of the condensed stiffness matrix and its computational cost,and parallel computing reduces the workload per process and enables the application of a parallel multigrid solver.Besides,several techniques,such as matrix-free implementation,direct condensation of uniform coarse elements,and adjusting computational resource limits,have been developed to enhance computational efficiency.The weak scaling efficiency,strong scaling speedup,and maximum achievable efficiency of the proposed method are validated across multiple numerical examples,showing significant improvement in the tractable problem size and solution efficiency compared to traditional topology optimization algorithms. 展开更多
关键词 Topology optimization Large-scale Problem independent machine learning(PIML) Parallel computing
暂未订购 下载PDF
A 32‑channel charge‑sensitive amplifier for delay‑line readout of parallel plate avalanche counter array 认领 引用
11
作者 Yue‑Zhao Zhang Peng Ma +8 位作者 Zhuang‑Yu Lin Zhen‑Fei Tan Xing‑Chi Han Chen Liu Shuo Wang Da‑Peng Sun Zhi‑Quan Li En‑Hong Wang Shou‑Yu Wang 《Nuclear Science and Techniques》 SCIE EI CAS CSCD 2026年第1期164-179,共16页
A 32-channel charge-sensitive amplifier(CSA)is designed for fast timing in the delay-line readout of a parallel plate avalanche counter(PPAC)array.It is realized on a PCB with operational amplifiers and other discrete... A 32-channel charge-sensitive amplifier(CSA)is designed for fast timing in the delay-line readout of a parallel plate avalanche counter(PPAC)array.It is realized on a PCB with operational amplifiers and other discrete components.Each channel consists of an integrator,a pole-zero cancellation net,and a linear amplification stage,which can be adapted to accommodate either positive or negative input signals.The RMS equivalent input noise charges are 3.3 fC,the conversion gains are approximately±2 mV∕fC,and the intrinsic time resolution reaches 32 ps.In the prototype PPAC application,the CSA performs as well as the commercial FTA820A amplifier,providing a position resolution as good as 0.17 mm,and exhibiting reliable stability during several hours of continuous data acquisition. 展开更多
关键词 Charge-sensitive amplifier Fast timing Parallel plate avalanche counter Delay-line Discrete components
暂未订购 下载PDF
Stiffness evaluation and experimental test of a novel redundantly actuated parallel machining robot 认领 引用
12
作者 Hanliang Fang Jian Wang +2 位作者 Shuyi Ge Fufu Yang Jun Zhang 《Chinese Journal of Mechanical Engineering》 SCIE EI CAS CSCD 2026年第1期413-422,共10页
Parallel machining robot is a new type of robotized equipment for high-efficiency machining structural com-ponents with complex geometries.Terminal rigidity is of great importance index for such type of equipment,whic... Parallel machining robot is a new type of robotized equipment for high-efficiency machining structural com-ponents with complex geometries.Terminal rigidity is of great importance index for such type of equipment,which affects their load capacity and working accuracy.Before a parallel machining robot can be used for heavy-load and high-efficiency machining,its terminal rigidity should be evaluated systematically.The present study is to quantitatively reveal the stiffness properties of a previously invented Z4 redundantly actuated parallel ma-chining robot(RAPMR).For this purpose,two critical issues,i.e.,stiffness modelling and index construction,are clarified to carry out stiffness evaluation of the Z4 RAPMR.Firstly,drawing on the screw theory,a semi-analytic stiffness model of the proposed RAPMR is established at a component level.Secondly,a set of virtual work-based stiffness indices is constructed to evaluate the terminal rigidity of parallel robots.Those indices have a consistent physical unit in describing linear and angular terminal rigidity.With these indices,the local and the global stiffness performance of the Z4 RAPMR are predicted.Thirdly,a laboratory prototype of the proposed RAPMR is fabricated.And the experimental test is performed to verify the correctness of the established stiffness model.The present work is expected to provide fundamental information for further light-weight design and rigidity enhancement. 展开更多
关键词 Parallel machining robot Redundantly actuated Terminal rigidity Stiffness model Experimental test
暂未订购 下载PDF
Symmetry Breaking in Parallel 1-K Sorption Coolers and Passive Suppression Strategy 认领 引用
13
作者 Lihao Lu Yan Lu +2 位作者 Zhenhua Jiang Shaoshuai Liu Yinong Wu 《Frontiers in Heat and Mass Transfer》 EI CAS 2026年第3期105-122,共18页
Sub-Kelvin cooling technology is a critical prerequisite for high-sensitivity detection in deep space exploration and quantum computing.Operating identical sorption coolers in parallel is a common engineering approach... Sub-Kelvin cooling technology is a critical prerequisite for high-sensitivity detection in deep space exploration and quantum computing.Operating identical sorption coolers in parallel is a common engineering approach to enhance cooling capacity and extend hold time for these cryogenic platforms.However,this study reports an unexpected"symmetry breaking"phenomenon observed in a parallel Helium-4 sorption cooling system where the cold heads are connected via Oxygen-Free High Thermal Conductivity(OFHC)copper linkages.Instead of the expected uniform load sharing,the system spontaneously evolves into an asymmetric"quasi-series"operational mode.In this state,one cooler preferentially consumes its liquid helium inventory while the other remains dormant,significantly reducing system efficiency.To elucidate the underlying physics,a transient thermal-fluidic resistance network model was developed and validated against experimental data obtained from a dual-cooler test rig pre-cooled by a G-M cryocooler.Theoretical analysis reveals that this thermal locking originates from a positive feedback loop driven by the temperature-dependent thermal conductivity of the copper straps.Experimental results further demonstrate that system stability degrades significantly with increasing thermal load,with the synchronization ratio dropping from 75.3%at 0 mW to 51.3%at 3 mW.This indicates that at higher temperatures,the destabilizing gain of the thermal link overwhelms the restoring stiffness of the sorption mechanism.To address this intrinsic instability,a passive suppression strategy using a series"Ballast Thermal Resistance"is proposed.Numerical optimization identifies a critical resistance value of approximately 10 K/W,which effectively dampens the positive feedback and restores the synchronization ratio to over 95%with a negligible thermal penalty of less than 20 mK.These findings provide a theoretical basis and practical design guidelines for the stabilization of multi-cooler cryogenic networks. 展开更多
关键词 Sorption cooler parallel system symmetry breaking thermal locking passive stabilization ballast thermal resistance
暂未订购 下载PDF
A Deep Reinforcement Learning-Based Partitioning Method for Power System Parallel Restoration 认领 引用
14
作者 Changcheng Li Weimeng Chang +1 位作者 Dahai Zhang Jinghan He 《Energy Engineering》 EI 2026年第1期243-264,共22页
Effective partitioning is crucial for enabling parallel restoration of power systems after blackouts.This paper proposes a novel partitioning method based on deep reinforcement learning.First,the partitioning decision... Effective partitioning is crucial for enabling parallel restoration of power systems after blackouts.This paper proposes a novel partitioning method based on deep reinforcement learning.First,the partitioning decision process is formulated as a Markov decision process(MDP)model to maximize the modularity.Corresponding key partitioning constraints on parallel restoration are considered.Second,based on the partitioning objective and constraints,the reward function of the partitioning MDP model is set by adopting a relative deviation normalization scheme to reduce mutual interference between the reward and penalty in the reward function.The soft bonus scaling mechanism is introduced to mitigate overestimation caused by abrupt jumps in the reward.Then,the deep Q network method is applied to solve the partitioning MDP model and generate partitioning schemes.Two experience replay buffers are employed to speed up the training process of the method.Finally,case studies on the IEEE 39-bus test system demonstrate that the proposed method can generate a high-modularity partitioning result that meets all key partitioning constraints,thereby improving the parallelism and reliability of the restoration process.Moreover,simulation results demonstrate that an appropriate discount factor is crucial for ensuring both the convergence speed and the stability of the partitioning training. 展开更多
关键词 Partitioning method parallel restoration deep reinforcement learning experience replay buffer partitioning modularity
暂未订购 下载PDF
A Competitive Parallel Animated Oat Optimization Algorithm for Reversible Digital Watermarking 认领 引用
15
作者 Shu-Chuan Chu Libin Fu Jeng-Shyang Pan 《Computer Modeling in Engineering & Sciences》 SCIE EI 2026年第7期1209-1238,共30页
The Animated Oat Optimization Algorithm(AOO)is a novel evolutionary algorithm inspired by the behavior of animated oats.This paper proposes a Competitive Parallel Animated Oat Optimization Algorithm(CPAOO)comprising t... The Animated Oat Optimization Algorithm(AOO)is a novel evolutionary algorithm inspired by the behavior of animated oats.This paper proposes a Competitive Parallel Animated Oat Optimization Algorithm(CPAOO)comprising two components.First,a parallel strategy is employed in which inter-subpopulation communication is triggered at predefined iteration thresholds to balance exploration and exploitation.Second,a grouped competition strategy with incentive mechanisms is introduced,enabling the prioritized evolution of superior individuals to enhance the algorithm’s efficiency.Furthermore,building on the Prediction Error Expansion(PEE)algorithm,this paper proposes a Dual-Layer PEE(DLPEE)algorithm for reversible digital watermarking.Based on differences in pixel values around embedding points,image blocks are classified as either smooth or textured regions.The CPAOO algorithm is used to optimize the weights of the pixel predictor and to prioritize embedding secret information in smooth blocks.This approach enhances both the embedding capacity and the invisibility of the watermarked data.Experimental results demonstrate that the proposed methods achieve satisfactory performance. 展开更多
关键词 Evolutionary algorithm parallel strategy communication strategy grouped competition strategy reversible digital watermarking
暂未订购 下载PDF
An Eulerian-Lagrangian parallel algorithm for simulation of particle-laden turbulent flows 认领 引用 被引量：1
16
作者 Harshal P.Mahamure Deekshith I.Poojary +1 位作者 Vagesh D.Narasimhamurthy Lihao Zhao 《Acta Mechanica Sinica》 SCIE EI CAS CSCD 2026年第1期15-34,共20页
This paper presents an Eulerian-Lagrangian algorithm for direct numerical simulation(DNS)of particle-laden flows.The algorithm is applicable to perform simulations of dilute suspensions of small inertial particles in ... This paper presents an Eulerian-Lagrangian algorithm for direct numerical simulation(DNS)of particle-laden flows.The algorithm is applicable to perform simulations of dilute suspensions of small inertial particles in turbulent carrier flow.The Eulerian framework numerically resolves turbulent carrier flow using a parallelized,finite-volume DNS solver on a staggered Cartesian grid.Particles are tracked using a point-particle method utilizing a Lagrangian particle tracking(LPT)algorithm.The proposed Eulerian-Lagrangian algorithm is validated using an inertial particle-laden turbulent channel flow for different Stokes number cases.The particle concentration profiles and higher-order statistics of the carrier and dispersed phases agree well with the benchmark results.We investigated the effect of fluid velocity interpolation and numerical integration schemes of particle tracking algorithms on particle dispersion statistics.The suitability of fluid velocity interpolation schemes for predicting the particle dispersion statistics is discussed in the framework of the particle tracking algorithm coupled to the finite-volume solver.In addition,we present parallelization strategies implemented in the algorithm and evaluate their parallel performance. 展开更多
关键词 DNS Eulerian-Lagrangian Particle tracking algorithm Point-particle Parallel software
暂未订购 下载PDF
Research on Ultra-Short-Term Photovoltaic Power Forecasting Based on Parallel Architecture TCN-BiLSTM with Temporal-Spatial Attention Mechanism 认领 引用 被引量：1
17
作者 Hongbo Sun Xingyu Jiang +4 位作者 Wenyao Sun Yi Zhao Jifeng Cheng Xiaoyi Qian Guo Wang 《Energy Engineering》 EI 2026年第4期303-320,共18页
The accuracy of photovoltaic(PV)power prediction is significantly influenced by meteorological and environmental factors.To enhance ultra-short-term forecasting precision,this paper proposes an interpretable feedback ... The accuracy of photovoltaic(PV)power prediction is significantly influenced by meteorological and environmental factors.To enhance ultra-short-term forecasting precision,this paper proposes an interpretable feedback prediction method based on a parallel dual-stream Temporal Convolutional Network-Bidirectional Long Short-Term Memory(TCN-BiLSTM)architecture incorporating a spatiotemporal attention mechanism.Firstly,during data preprocessing,the optimal historical time window is determined through autocorrelation analysis while highly correlated features are selected as model inputs using Pearson correlation coefficients.Subsequently,a parallel dual-stream TCN-BiLSTM model is constructed where the TCN branch extracts localized transient features and the BiLSTM branch captures long-term periodic patterns,with spatiotemporal attention dynamically weighting spatiotemporal dependencies.Finally,Shapley Additive explanations(SHAP)additive analysis quantifies feature contribution rates and provides optimization feedback to the model.Validation using operational data from a PV power station in Northeast China demonstrates that compared to conventional deep learning models,the proposed method achieves a 17.6%reduction in root mean square error(RMSE),a 5.4%decrease in training time consumption,and a 4.78%improvement in continuous ranked probability score(CRPS),exhibiting significant advantages in both prediction accuracy and generalization capability.This approach enhances the application effectiveness of ultra-short-term PV power forecasting while simultaneously improving prediction accuracy and computational efficiency. 展开更多
关键词 Ultra-short-term forecasting temporal convolutional network bidirectional long short-term memory parallel dual-stream architecture temporal-spatial attention SHAP contribution analysis
暂未订购 下载PDF
Multi-task hierarchical network for semantic understanding of air traffic controller-pilot communication 认领 引用
18
作者 Xiaoxiao ZHANG Qihan DENG +4 位作者 Yang YANG Shengsheng QIAN Yi HUI Yanbo ZHU Kaiquan CAI 《Chinese Journal of Aeronautics》 SCIE EI CAS CSCD 2026年第3期499-515,共17页
Flight situational awareness in civil aviation relies on the semantic understanding of both the key details and the full picture from the Air Traffic Controller(ATCo)and pilot communication.This paper proposes a novel... Flight situational awareness in civil aviation relies on the semantic understanding of both the key details and the full picture from the Air Traffic Controller(ATCo)and pilot communication.This paper proposes a novel end-to-end Multi-Task Hierarchical Network(MTHN)for automatically understanding ATCo-pilot communication,handling slot filling,role detection,and intent recognition at different levels while adaptively integrating them.Specifically,we introduce a wordbased knowledge-masked slot distillation module that constructs an ATC knowledge base to dynamically mask keywords during teacher-student distillation.Considering the distinct intent differences between ATCos and pilots,we design a sentence-based role-aware intent attention module that extracts role label space vectors as context to enrich intent representations.To exploit the complementarity across different semantic levels in ATCo-pilot communication,we explicitly develop an adaptive bi-interaction flow module that dynamically explores semantic dependencies among tasks.Extensive experiments on real-world datasets collected in China show the superior performance of MTHN,compared to state-of-the-art baselines in both general natural language understanding and ATC-specific text processing.Our results highlight that MTHN achieves 99.26%,97.25%,and 96.22%accuracy across key slots,as well as 96.59%accuracy in speaker role classification.Moreover,it can perceive multi-label deep intents behind sentences.These analytical findings demonstrate the potential to reduce human errors in high-concurrency ATCo-pilot interactions under dense operational conditions. 展开更多
关键词 Air traffic control Attention mechanism Hierarchical modeling Multi-task learning Semantic understanding
暂未订购 下载PDF
A surface emphasized multi-task learning framework for surface property predictions:A case study of magnesium intermetallics 认领 引用
19
作者 Gaoning Shi Yaowei Wang +3 位作者 Kun Yang Yuan Qiu Hong Zhu Xiaoqin Zeng 《Journal of Magnesium and Alloys》 SCIE EI CAS CSCD 2026年第1期216-227,共12页
Surface properties of crystals are critical in many fields,including electrochemistry and photoelectronics,the efficient prediction of which can expedite the design and optimization of catalysts,batteries,alloys etc.H... Surface properties of crystals are critical in many fields,including electrochemistry and photoelectronics,the efficient prediction of which can expedite the design and optimization of catalysts,batteries,alloys etc.However,we are still far from realizing this vision due to the rarity of surface property-related databases,especially for multicomponent compounds,due to the large sample spaces and limited computing resources.In this work,we present a surface emphasized multi-task crystal graph convolutional neural network(SEM-CGCNN)to predict multiple surface properties simultaneously from crystal structures.The model is evaluated on a dataset of 3526 surface energies and work functions of binary magnesium intermetallics obtained through first-principles calculations,and obvious improvements are observed both in efficiency and accuracy over the original CGCNN model.By transferring the pre-trained model to the datasets of pure metals and other intermetallics,the fine-tuned SEM-CGCNN outperforms learning from scratch and can be further applied to other surface properties and materials systems.This study could be a paradigm for the end-to-end mapping of atomic structures to anisotropic surface properties of crystals,which provides an efficient framework to understand and screen materials with desired surface characteristics. 展开更多
关键词 Graph neural networks Multi-task learning Surface energy Work function Intermetallic compounds Mg alloy
暂未订购 下载PDF
A Subdomain-Based GPU Parallel Scheme for Accelerating Perdynamics Modeling with Reduced Graphics Memory 认领 引用
20
作者 Zuokun Yang Jun Li +1 位作者 Xin Lai Lisheng Liu 《Computer Modeling in Engineering & Sciences》 SCIE EI 2026年第1期256-285,共30页
Peridynamics(PD)demonstrates unique advantages in addressing fracture problems,however,its nonlocality and meshfree discretization result in high computational and storage costs.Moreover,in its engineering application... Peridynamics(PD)demonstrates unique advantages in addressing fracture problems,however,its nonlocality and meshfree discretization result in high computational and storage costs.Moreover,in its engineering applications,the computational scale of classical GPU parallel schemes is often limited by the finite graphics memory of GPU devices.In the present study,we develop an efficient particle information management strategy based on the cell-linked list method and on this basis propose a subdomain-based GPU parallel scheme,which exhibits outstanding acceleration performance in specific compute kernels while significantly reducing graphics memory usage.Compared to the classical parallel scheme,the cell-linked list method facilitates efficient management of particle information within subdomains,enabling the proposed parallel scheme to effectively reduce graphics memory usage by optimizing the size and number of subdomains while significantly improving the speed of neighbor search.As demonstrated in PD examples,the proposed parallel scheme enhances the neighbor search efficiency dramatically and achieves a significant speedup relative to serial programs.For instance,without considering the time of data transmission,the proposed scheme achieves a remarkable speedup of nearly 1076.8×in one test case,due to its excellent computational efficiency in the neighbor search.Additionally,for 2D and 3D PD models with tens of millions of particles,the graphics memory usage can be reduced up to 83.6%and 85.9%,respectively.Therefore,this subdomain-based GPU parallel scheme effectively avoids graphics memory shortages while significantly improving the computational efficiency,providing new insights into studying more complex large-scale problems. 展开更多
关键词 Peridynamics GPU CUDA parallel computing cell-linked list
暂未订购 下载PDF
已选0条
引用分析
参考文献
引证文献
 统计分析
检索结果
已选文献
上一页 1 2 … 250 下一页 到第 页
在线咨询 使用帮助 返回顶部 意见反馈