Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not ...Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not in a time-efficient manner.The time efficiency becomes even worse when the flows to be classified are sampled by flow entry timeout over Software-Defined Networks(SDNs)to achieve a better resource efficiency.This paper addresses this situation by combining co-training and Reinforcement Learning(RL)to enable a closed-loop classification approach that divides the entire classification process into episodes,each involving two elephant models.One predicts elephants and is retrained by a selection of flows automatically labeled online by the other.RL is used to formulate a reward function that estimates the values of the possible actions based on the current states of both models and further adjusts the ratio of flows to be labeled in each phase.Extensive evaluation based on real traffic traces shows that the proposed approach can stably predict elephants using the packets received in the first 10% of their lifetime with an accuracy of over 80%,and using only about 10% more control channel bandwidth than the baseline over the evolved SDNs.展开更多
Rational allocation of water flow energy in river networks is essential to addressing water-related issues in river network areas.However,current methods of calculating the spatiotemporal distribution of flow energy i...Rational allocation of water flow energy in river networks is essential to addressing water-related issues in river network areas.However,current methods of calculating the spatiotemporal distribution of flow energy in river networks lack precision and efficiency.This paper introduces a novel hydrodynamic representation,the energy flow rate,defined as the product of the flow rate and kinetic energy head,to quantify the kinetic energy stored and transported in river networks.A linear equation system for the energy flow rate in a river network has been theoretically derived,enabling rapid calculations under steady flow conditions.A simplified equation is proposed to describe the exponential decay of the energy flow rate,accompanied by potential energy conversion.The coefficients in the linear equation system are determined using control equations at flow confluence and diversion nodes.This study provides foundational insights that can be used to develop new hydrodynamic modeling strategies to regulate water flow energy and achieve coordinated management of water-related issues in river networks.展开更多
Zero Trust Network(ZTN)enhances network security through strict authentication and access control.However,in the ZTN,optimizing flow control to improve the quality of service is still facing challenges.Software Define...Zero Trust Network(ZTN)enhances network security through strict authentication and access control.However,in the ZTN,optimizing flow control to improve the quality of service is still facing challenges.Software Defined Network(SDN)provides solutions through centralized control and dynamic resource allocation,but the existing scheduling methods based on Deep Reinforcement Learning(DRL)are insufficient in terms of convergence speed and dynamic optimization capability.To solve these problems,this paper proposes DRL-AMIR,which is an efficient flow scheduling method for software defined ZTN.This method constructs a flow scheduling optimization model that comprehensively considers service delay,bandwidth occupation,and path hops.Additionally,it balances the differentiated requirements of delay-critical K-flows,bandwidth-intensive D-flows,and background B-flows through adaptiveweighting.Theproposed framework employs a customized state space comprising node labels,link bandwidth,delaymetrics,and path length.It incorporates an action space derived fromnode weights and a hybrid reward function that integrates both single-step and multi-step excitation mechanisms.Based on these components,a hierarchical architecture is designed,effectively integrating the data plane,control plane,and knowledge plane.In particular,the adaptive expert mechanism is introduced,which triggers the shortest path algorithm in the training process to accelerate convergence,reduce trial and error costs,and maintain stability.Experiments across diverse real-world network topologies demonstrate that DRL-AMIR achieves a 15–20%reduction in K-flow transmission delays,a 10–15%improvement in link bandwidth utilization compared to SPR,QoSR,and DRSIR,and a 30%faster convergence speed via adaptive expert mechanisms.展开更多
Accurate traffic flow prediction has a profound impact on modern traffic management. Traffic flow has complex spatial-temporal correlations and periodicity, which poses difficulties for precise prediction. To address ...Accurate traffic flow prediction has a profound impact on modern traffic management. Traffic flow has complex spatial-temporal correlations and periodicity, which poses difficulties for precise prediction. To address this problem, a Multi-head Self-attention and Spatial-Temporal Graph Convolutional Network (MSSTGCN) for multiscale traffic flow prediction is proposed. Firstly, to capture the hidden traffic periodicity of traffic flow, traffic flow is divided into three kinds of periods, including hourly, daily, and weekly data. Secondly, a graph attention residual layer is constructed to learn the global spatial features across regions. Local spatial-temporal dependence is captured by using a T-GCN module. Thirdly, a transformer layer is introduced to learn the long-term dependence in time. A position embedding mechanism is introduced to label position information for all traffic sequences. Thus, this multi-head self-attention mechanism can recognize the sequence order and allocate weights for different time nodes. Experimental results on four real-world datasets show that the MSSTGCN performs better than the baseline methods and can be successfully adapted to traffic prediction tasks.展开更多
The numerical calculation method has greatly promoted the process of optimal design of scramjet,but it still needs extremely heavy calculation for the model with complex thermochemical reaction.Data-driven deep learni...The numerical calculation method has greatly promoted the process of optimal design of scramjet,but it still needs extremely heavy calculation for the model with complex thermochemical reaction.Data-driven deep learning relies heavily on a large amount of data in the face of complex nonlinear features.Therefore,combining“data-driven model”and“Navier-Stokes equation”,an intelligent prediction model of supersonic combustion flow process is constructed.This algorithm integrates the theory priors of combustion flow into the neural network model,and uses convolutional grouping and rearrangement to reduce the feature redundancy calculation,so as to achieve high-precision and high-efficiency prediction of velocity,density,pressure and temperature fields.This study makes a comprehensive comparison from two aspects of performance and efficiency.Unsteady scramjet multi-physical field dataset is constructed under different incoming Mach number conditions.The experimental results show that compared with other methods,the proposed algorithm can achieve the maximum Peak Signal-to-Noise Ratio(PSNR)improvement of 38.75%and Learned Perceptual Image Patch Similarity(LPIPS)improvement of 68.13%in predicting the average quality of images,and the computational cost of the model is reduced by 30.36%compared with other models.In addition,the high model can also effectively predict the unknown incoming flow condition.展开更多
The presence of circles in the network maximum flow problem increases the complexity of the preflow algorithm.This study proposes a novel two-stage preflow algorithm to address this issue.First,this study proves that ...The presence of circles in the network maximum flow problem increases the complexity of the preflow algorithm.This study proposes a novel two-stage preflow algorithm to address this issue.First,this study proves that at least one zero-flow arc must be present when the flow of the network reaches its maximum value.This result indicates that the maximum flow of the network will remain constant if a zero-flow arc within a circle is removed;therefore,the maximum flow of each network without circles can be calculated.The first stage involves identifying the zero-flow arc in the circle when the network flow reaches its maximum.The second stage aims to remove the zero-flow arc identified and modified in the first stage,thereby producing a new network without circles.The maximum flow of the original looped network can be obtained by solving the maximum flow of the newly generated acyclic network.Finally,an example is provided to demonstrate the validity and feasibility of this algorithm.This algorithm not only improves computational efficiency but also provides new perspectives and tools for solving similar network optimization problems.展开更多
This research introduces a spectrum-based physics-informed neural network(SP-PINN)model to significantly improve the accuracy of calculation of two-phase flow parameters,surpassing existing methods that have limitatio...This research introduces a spectrum-based physics-informed neural network(SP-PINN)model to significantly improve the accuracy of calculation of two-phase flow parameters,surpassing existing methods that have limitations in global and continuous data sampling.SP-PINNs address the challenges of traditional methods in terms of continuous sampling by integrating the spectral analysis and pressure correction into the Navier-Stokes(N-S)equations,enhancing the predictive accuracy especially in critical regions like gas-phase boundaries and velocity peaks.The novel introduction of a pressure-correction module within SP-PINNs mitigates prediction errors,achieving a substantial reduction to 1‰compared with the conventional physics-informed neural network(PINN)approaches.Experimental applications validate the model’s ability to accurately and rapidly predict flow parameters with different sampling time intervals,with the computation time of predicting unsampled data less than 0.01 s.Such advancements signify a 100-fold improvement over traditional DNS calculations,underscoring the model’s potential in the real-time calculation and analysis of multiphase flow dynamics.展开更多
With the widespread implementation of distributed generation(DG)and the integration of soft open point(SOP)into the distribution network(DN),the latter is steadily transitioning into a flexible distribution network(FD...With the widespread implementation of distributed generation(DG)and the integration of soft open point(SOP)into the distribution network(DN),the latter is steadily transitioning into a flexible distribution network(FDN),the calculation of carbon flow distribution in FDN is more difficult.To this end,this study constructs a model for low-carbon optimal operations within the FDN on the basis of enhanced carbon emission flow(CEF).First,the carbon emission characteristics of FDNs are scrutinized and an improved method for calculating carbon flow within these networks is proposed.Subsequently,a model for optimizing low-carbon operations within FDNs is formulated based on the refined CEF,which merges the specificities of DG and intelligent SOP.Finally,this model is scrutinized using an upgraded IEEE 33-node distribution system,a comparative analysis of the cases reveals that when DG and SOP are operated in a coordinated manner in the FDN,with the cost of electricity generation was reduced by 40.63 percent and the cost of carbon emissions by 10.18 percent.The findings indicate that the judicious optimization of areas exhibiting higher carbon flow rates can effectively enhance the economic efficiency of DN operations and curtail the carbon emissions of the overall network.展开更多
With the rapid development of deep learning neural networks,new solutions have emerged for addressing fluid flow problems in porous media.Combining data-driven approaches with physical constraints has become a hot res...With the rapid development of deep learning neural networks,new solutions have emerged for addressing fluid flow problems in porous media.Combining data-driven approaches with physical constraints has become a hot research direction,with physics-informed neural networks(PINNs) being the most popular hybrid model.PINNs have gained widespread attention in subsurface fluid flow simulations due to their low computational resource requirements,fast training speeds,strong generalization capabilities,and broad applicability.Despite success in homogeneous settings,standard PINNs face challenges in accurately calculating flux between irregular Eulerian cells with disparate properties and capturing global field influences on local cells.This limits their suitability for heterogeneous reservoirs and the irregular Eulerian grids frequently used in reservoir.To address these challenges,this study proposes a physics-informed graph neural network(PIGNN) model.The PIGNN model treats the entire field as a whole,integrating information from neighboring grids and physical laws into the solution for the target grid,thereby improving the accuracy of solving partial differential equations in heterogeneous and Eulerian irregular grids.The optimized model was applied to pressure field prediction in a spatially heterogeneous reservoir,achieving an average L_(2) error and R_(2) score of 6.710×10^(-4)and 0.998,respectively,which confirms the effectiveness of model.Compared to the conventional PINN model,the average L_(2) error was reduced by 76.93%,the average R_(2) score increased by 3.56%.Moreover,evaluating robustness,training the PIGNN model using only 54% and 76% of the original data yielded average relative L_(2) error reductions of 58.63% and 56.22%,respectively,compared to the PINN model.These results confirm the superior performance of this approach compared to PINN.展开更多
Two actual rocks drilled from a typical ultra-deep hydrocarbon reservoir in the Tarim Basin are selected to conduct in-situ stress-loading micro-focus CT scanning experiments.The gray images of rock microstructure at ...Two actual rocks drilled from a typical ultra-deep hydrocarbon reservoir in the Tarim Basin are selected to conduct in-situ stress-loading micro-focus CT scanning experiments.The gray images of rock microstructure at different stress loading stages are obtained.The U-Net fully convolutional neural network is utilized to achieve fine semantic segmentation of rock skeleton,pore space,and microfractures based on CT slice images of deep rocks.The three-dimensional digital rock models of deformed multiscale fractured-porous media at different stress loading stages are thereafter reconstructed,and the equivalent fracture-pore network models are finally extracted to explore the underlying mechanisms of gas-water two-phase flow at the pore-scale.Results indicate that,in the process of insitu stress loading,both the deep rocks have experienced three stages:linear elastic deformation,nonlinear plastic deformation,and shear failure.The micro-mechanical behavior greatly affects the dynamic deformation of rock microstructure and gas-water two-phase flow.In the linear elastic deformation stage,with the increase in in-situ stress,both the deep rocks are gradually compacted,leading to decreases in average pore radius,pore throat ratio,tortuosity,and water-phase relative permeability,while the coordination number nearly remains unchanged.In the plastic deformation stage,the synergistic influence of rock compaction and existence of micro-fractures typically exert a great effect on pore-throat topological properties and gas-water relative permeability.In the shear failure stage,due to the generation and propagation of micro-fractures inside the deep rock,the topological connectivity becomes better,fluid flow paths increase,and flow conductivity is promoted,thus leading to sharp increases in average pore radius and coordination number,rapid decreases in pore throat ratio and tortuosity,as well as remarkable improvement in relative permeability of gas phase and waterphase.展开更多
Precise traffic flow forecasting is essential for mitigating urban traffic congestion.However,it is difficult for existing methods to adequately capture the dynamic spatio-temporal characteristics and multiscale tempo...Precise traffic flow forecasting is essential for mitigating urban traffic congestion.However,it is difficult for existing methods to adequately capture the dynamic spatio-temporal characteristics and multiscale temporal dependencies of traffic flow.A traffic flow prediction model with multiscale temporal awareness and graph diffusion attention networks(MT-GDAN)is proposed to address these issues.Specifically,a graph diffusion attention module is constructed,which dynamically adjusts and calculates the weights of neighboring nodes in the graph structure using a random graph attention network(GAT)and captures the spatial characteristics of hidden nodes through an adaptive adjacency matrix,thus better exploiting the dynamic spatio-temporal properties of traffic flow.Secondly,a multiscale isometric convolutional network and bi-level routing attention are used to construct a multiscale temporal awareness module.The former extracts local information of traffic flow segments by convolution with different sizes of convolution kernels and then introduces isometric convolution to obtain the global temporal relationship between local features of traffic flow segments;the latter filters irrelevant spatio-temporal features at a coarse regional level and focuses locally on key points to more accurately capture the multiscale temporal dependencies of traffic flows.Experimental results reveal that the MT-GDAN model surpasses the mainstream baseline model in terms of forecasting accuracy and exhibits good prediction performance.展开更多
Short⁃term traffic flow prediction plays a crucial role in the planning of intelligent transportation systems.Nowadays,there is a large amount of traffic flow data generated from the monitoring devices of urban road n...Short⁃term traffic flow prediction plays a crucial role in the planning of intelligent transportation systems.Nowadays,there is a large amount of traffic flow data generated from the monitoring devices of urban road networks,which contains road network traffic information with high application value.In this study,an improved spatio⁃temporal attention transformer model(ISTA⁃transformer model)is proposed to provide a more accurate method for predicting multi⁃step short⁃term traffic flow based on monitoring data.By embedding a temporal attention layer and a spatial attention layer in the model,the model learns the relationship between traffic flows at different time intervals and different geographic locations,and realizes more accurate multi⁃step short⁃time flow prediction.Finally,we validate the superiority of the model with monitoring data spanning 15 days from 620 monitoring points in Qingdao,China.In the four time steps of prediction,the MAPE(Mean Absolute Percentage Error)values of ISTA⁃transformers prediction results are 0.22,0.29,0.37,and 0.38,respectively,and its prediction accuracy is usually better than that of six baseline models(Transformer,GRU,CNN,LSTM,Seq2Seq and LightGBM),which indicates that the proposed model in this paper always has a better ability to explain the prediction results with the time steps in the multi⁃step prediction.展开更多
The global clustering of inventive talent shapes innovation capacity and drives economic growth.For China,this process is especially crucial in sustaining its development momentum.This paper draws on data from the EPO...The global clustering of inventive talent shapes innovation capacity and drives economic growth.For China,this process is especially crucial in sustaining its development momentum.This paper draws on data from the EPO Worldwide Patent Statistical Database(PATSTAT)to extract global inventive talent mobility information and analyzes the spatial structural evolution of the global inventive talent flow network.The study finds that this network is undergoing a multi-polar transformation,characterized by the rising importance of a few central countries-such as the United States,Germany,and China-and the increasing marginalization of many peripheral countries.In response to this typical phenomenon,the paper constructs an endogenous migration model and conducts empirical testing using the Temporal Exponential Random Graph Model(TERGM).The results reveal several endogenous mechanisms driving global inventive talent flows,including reciprocity,path dependence,convergence effects,transitivity,and cyclic structures,all of which contribute to the network’s multi-polar trend.In addition,differences in regional industrial structures significantly influence talent mobility choices and are a decisive factor in the formation of poles within the multi-polar landscape.Based on these findings,it is suggested that efforts be made to foster two-way channels for talent exchange between China and other global innovation hubs,in order to enhance international collaboration and knowledge flow.We should aim to reduce the migration costs and institutional barriers faced by R&D personnel,thereby encouraging greater mobility of high-skilled talent.Furthermore,the government is advised to strategically leverage regional strengths in high-tech industries as a lever to capture competitive advantages in emerging technologies and products,ultimately strengthening the country’s position in the global innovation landscape.展开更多
Physics-informed neural networks(PINNs)have prevailed as differentiable simulators to investigate flow in porous media.Despite recent progress PINNs have achieved,practical geotechnical scenarios cannot be readily sim...Physics-informed neural networks(PINNs)have prevailed as differentiable simulators to investigate flow in porous media.Despite recent progress PINNs have achieved,practical geotechnical scenarios cannot be readily simulated because conventional PINNs fail in discontinuous heterogeneous porous media or multi-layer strata when labeled data are missing.This work aims to develop a universal network structure to encode the mass continuity equation and Darcy’s law without labeled data.The finite element approximation,which can decompose a complex heterogeneous domain into simpler ones,is adopted to build the differentiable network.Without conventional DNNs,physics-encoded finite element network(PEFEN)can avoid spectral bias and learn high-frequency functions with sharp/steep gradients.PEFEN rigorously encodes Dirichlet and Neumann boundary conditions without training.Benefiting from its discretized formulation,the discontinuous heterogeneous hydraulic conductivity is readily embedded into the network.Three typical cases are reproduced to corroborate PEFEN’s superior performance over conventional PINNs and the PINN with mixed formulation.PEFEN is sparse and demonstrated to be capable of dealing with heterogeneity with much fewer training iterations(less than 1/30)than the improved PINN with mixed formulation.Thus,PEFEN saves energy and contributes to low-carbon AI for science.The last two cases focus on common geotechnical settings of impermeable sheet pile in singlelayer and multi-layer strata.PEFEN solves these cases with high accuracy,circumventing costly labeled data,extra computational burden,and additional treatment.Thus,this study warrants the further development and application of PEFEN as a novel differentiable network in porous flow of practical geotechnical engineering.展开更多
Characteristics of the Internet traffic data flow are studied based on the chaos theory. A phase space that is isometric with the network dynamic system is reconstructed by using the single variable time series of a n...Characteristics of the Internet traffic data flow are studied based on the chaos theory. A phase space that is isometric with the network dynamic system is reconstructed by using the single variable time series of a network flow. Some parameters, such as the correlative dimension and the Lyapunov exponent are calculated, and the chaos characteristic is proved to exist in Internet traffic data flows. A neural network model is construct- ed based on radial basis function （RBF） to forecast actual Internet traffic data flow. Simulation results show that, compared with other forecasts of the forward-feedback neural network, the forecast of the RBF neural network based on the chaos theory has faster learning capacity and higher forecasting accuracy.展开更多
The explicit rate flow control mechanisms for ABR service are used to sharethe available bandwidth of a bottleneck link fairly and reasonably among many competitive users andto maintain the buffer queue length of a bo...The explicit rate flow control mechanisms for ABR service are used to sharethe available bandwidth of a bottleneck link fairly and reasonably among many competitive users andto maintain the buffer queue length of a bottleneck switch connected to the link at a desired levelin order to avoid and control congestion in ATM networks. However, designing effective flow controlmechanisms for the service is known to be difficult because of the variety of dynamic parametersinvolved such as available link bandwidth, burst of the traffic, the distances between ABR sourcesand switches. In this paper, we present a fuzzy explicit rate flow control mechanism for ABRservice. The mechanism has a simple structure and is robust in the sense that the mechanism'sstability is not sensitive to the change in the number of active virtual connections (VCs). Manysimulations show that this mechanism can not only effectively avoid network congestion, but alsoensure fair share of the bandwidth for all active VCs regardless of the number of hops theytraverse. Additionally, it has the advantages of fast convergence, low oscillation, and high linkbandwidth utilization.展开更多
Blockage is a kind of phenomenon frequently occurred in a transport network, in which the human beings are the moving subjects. The minimum flow of a network defined in this paper means the maximum flow quantity throu...Blockage is a kind of phenomenon frequently occurred in a transport network, in which the human beings are the moving subjects. The minimum flow of a network defined in this paper means the maximum flow quantity through the network in the seriously blocked situation. It is an important parameter in designing and operating a transport network, especially in an emergency evacuation network. A branch and bound method is presented to solve the minimum flow problem on the basis of the blocking flow theory and the algorithm and its application are illustrated by examples.展开更多
Blockage is a kind of phenomenon occurring frequently in modern transportation network. This paper deals with the research work on the blocking now in a network with the help of network flow theory. The blockage pheno...Blockage is a kind of phenomenon occurring frequently in modern transportation network. This paper deals with the research work on the blocking now in a network with the help of network flow theory. The blockage phenomena can be divided intO local blockage and network blockage. In this paper, which deals mainly with the latter, the fundamental concepts and definitions of network blocking flow, blocking outset are presented and the related theorems are proved. It is proved that the sufficient and necessary condition for the emergence of a blocking now in a network is the existence of the blocking outset. The necessary conditions for the existence of the blocking outset in a network are analysed and the characteristic cutset of blockage which reflects the all possible situation of blocking nows in the network is defined.In the last part of the paper the mathematical model of the minimum blocking now is developed and the solution to a small network is given.展开更多
Neuroimaging of cerebral glucose metabolism and blood flow is ideally suited to assay widely-distributed brain circuits as a result of local molecular events and behavioral modulation in the central nervous system. Wi...Neuroimaging of cerebral glucose metabolism and blood flow is ideally suited to assay widely-distributed brain circuits as a result of local molecular events and behavioral modulation in the central nervous system. With the progress in novel analytical methodology, this endeavor has succeeded in unraveling the mechanisms underlying a wide spectrum of neurodegenerative diseases. In particular, statistical brain mapping studies have made significant strides in describing the pathophysiology of Parkinson＇s disease （PD） and related disorders by providing signature biomarkers to determine the systemic abnormalities in brain function and evaluate disease progression, therapeutic responses, and clinical correlates in patients. In this article, we review the relevant clinical applications in patients in relation to healthy volunteers with a focus on the generation of unique spatial covariance patterns associated with the motor and cognitive symptoms underlying PD. These characteristic biomarkers can be potentially used not only to improve patient recruitment but also to predict outcomes in clinical trials.展开更多
Although the wireless network is widely used in many fields,its characteristics such as high bit error rate and broadcast links may block its development.Network coding is an artistic way to exploit its intrinsic char...Although the wireless network is widely used in many fields,its characteristics such as high bit error rate and broadcast links may block its development.Network coding is an artistic way to exploit its intrinsic characteristics to increase the network reliability.Some people research network coding schemes for inter-flow or intra-flow,each type with its own advantages and disadvantages.In this paper,we propose a new mechanism,called MM-NCOPE,which integrates the idea of inter-flow and intra-flow coding.On the one hand,MM-NCOPE utilizes random liner coding to encode the NCOPE packets while NCOPE is a sub-protocol for optimizing the COPE algorithm by iteration.In NCOPE,packets are automatically matched by size to be coded.As a result,it improves the coding gain in some level.On the other hand,we adopt the partial Acknowledgement retransmission scheme to achieve high compactness and robustness.ACK is an independent packet with the highest priority rather than a part of the data packets.Compared with existing works on opportunistic network coding,our approach ensures the reliability of wireless links and improves the coding gain.展开更多
基金supported by the National Natural Science Foundation of China(61962016)the Ministry of Science and Technology of China(G2022033002L)+1 种基金National Natural Science Foundation of Guangxi(2022JJA170057)Guangxi Education Department’s Project on Improving the Basic Research Ability of Young and Middleaged Teachers in Universities(2023ky0812,Research on Statistical Network Delay Predictions in Large-scale SDNs).
文摘Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not in a time-efficient manner.The time efficiency becomes even worse when the flows to be classified are sampled by flow entry timeout over Software-Defined Networks(SDNs)to achieve a better resource efficiency.This paper addresses this situation by combining co-training and Reinforcement Learning(RL)to enable a closed-loop classification approach that divides the entire classification process into episodes,each involving two elephant models.One predicts elephants and is retrained by a selection of flows automatically labeled online by the other.RL is used to formulate a reward function that estimates the values of the possible actions based on the current states of both models and further adjusts the ratio of flows to be labeled in each phase.Extensive evaluation based on real traffic traces shows that the proposed approach can stably predict elephants using the packets received in the first 10% of their lifetime with an accuracy of over 80%,and using only about 10% more control channel bandwidth than the baseline over the evolved SDNs.
基金supported by the National Natural Science Foundation of China(Grant No.U2340221)the National Key R&D Programof China(Grant No.2022YFC3202602)+1 种基金the NaturalScience Foundation of Jiangsu Province(Grant No.BK20230036)the 111 Project(Grant No.B17015)。
文摘Rational allocation of water flow energy in river networks is essential to addressing water-related issues in river network areas.However,current methods of calculating the spatiotemporal distribution of flow energy in river networks lack precision and efficiency.This paper introduces a novel hydrodynamic representation,the energy flow rate,defined as the product of the flow rate and kinetic energy head,to quantify the kinetic energy stored and transported in river networks.A linear equation system for the energy flow rate in a river network has been theoretically derived,enabling rapid calculations under steady flow conditions.A simplified equation is proposed to describe the exponential decay of the energy flow rate,accompanied by potential energy conversion.The coefficients in the linear equation system are determined using control equations at flow confluence and diversion nodes.This study provides foundational insights that can be used to develop new hydrodynamic modeling strategies to regulate water flow energy and achieve coordinated management of water-related issues in river networks.
基金supported in part by Scientific Research Fund of Zhejiang Provincial Education Department under Grant Y202351110in part by Huzhou Science and Technology Plan Project under Grant 2024YZ23+1 种基金in part by Research Fund of National Key Laboratory of Advanced Communication Networks under Grant SCX23641X004in part by Postgraduate Research and Innovation Project of Huzhou University under Grant 2024KYCX50.
文摘Zero Trust Network(ZTN)enhances network security through strict authentication and access control.However,in the ZTN,optimizing flow control to improve the quality of service is still facing challenges.Software Defined Network(SDN)provides solutions through centralized control and dynamic resource allocation,but the existing scheduling methods based on Deep Reinforcement Learning(DRL)are insufficient in terms of convergence speed and dynamic optimization capability.To solve these problems,this paper proposes DRL-AMIR,which is an efficient flow scheduling method for software defined ZTN.This method constructs a flow scheduling optimization model that comprehensively considers service delay,bandwidth occupation,and path hops.Additionally,it balances the differentiated requirements of delay-critical K-flows,bandwidth-intensive D-flows,and background B-flows through adaptiveweighting.Theproposed framework employs a customized state space comprising node labels,link bandwidth,delaymetrics,and path length.It incorporates an action space derived fromnode weights and a hybrid reward function that integrates both single-step and multi-step excitation mechanisms.Based on these components,a hierarchical architecture is designed,effectively integrating the data plane,control plane,and knowledge plane.In particular,the adaptive expert mechanism is introduced,which triggers the shortest path algorithm in the training process to accelerate convergence,reduce trial and error costs,and maintain stability.Experiments across diverse real-world network topologies demonstrate that DRL-AMIR achieves a 15–20%reduction in K-flow transmission delays,a 10–15%improvement in link bandwidth utilization compared to SPR,QoSR,and DRSIR,and a 30%faster convergence speed via adaptive expert mechanisms.
基金supported by the National Natural Science Foundation of China(Grant Nos.62472149,62376089,62202147)Hubei Provincial Science and Technology Plan Project(2023BCB04100).
文摘Accurate traffic flow prediction has a profound impact on modern traffic management. Traffic flow has complex spatial-temporal correlations and periodicity, which poses difficulties for precise prediction. To address this problem, a Multi-head Self-attention and Spatial-Temporal Graph Convolutional Network (MSSTGCN) for multiscale traffic flow prediction is proposed. Firstly, to capture the hidden traffic periodicity of traffic flow, traffic flow is divided into three kinds of periods, including hourly, daily, and weekly data. Secondly, a graph attention residual layer is constructed to learn the global spatial features across regions. Local spatial-temporal dependence is captured by using a T-GCN module. Thirdly, a transformer layer is introduced to learn the long-term dependence in time. A position embedding mechanism is introduced to label position information for all traffic sequences. Thus, this multi-head self-attention mechanism can recognize the sequence order and allocate weights for different time nodes. Experimental results on four real-world datasets show that the MSSTGCN performs better than the baseline methods and can be successfully adapted to traffic prediction tasks.
基金supported by the Program of Key Projects of Foundation Strengthening Plan,China(No.2022-JCJQ-ZD114-12-03)the Graduate Student Innovation Fund Lighthouse Program of Southwest University of Science and Technology,China(No.24ycx3009)。
文摘The numerical calculation method has greatly promoted the process of optimal design of scramjet,but it still needs extremely heavy calculation for the model with complex thermochemical reaction.Data-driven deep learning relies heavily on a large amount of data in the face of complex nonlinear features.Therefore,combining“data-driven model”and“Navier-Stokes equation”,an intelligent prediction model of supersonic combustion flow process is constructed.This algorithm integrates the theory priors of combustion flow into the neural network model,and uses convolutional grouping and rearrangement to reduce the feature redundancy calculation,so as to achieve high-precision and high-efficiency prediction of velocity,density,pressure and temperature fields.This study makes a comprehensive comparison from two aspects of performance and efficiency.Unsteady scramjet multi-physical field dataset is constructed under different incoming Mach number conditions.The experimental results show that compared with other methods,the proposed algorithm can achieve the maximum Peak Signal-to-Noise Ratio(PSNR)improvement of 38.75%and Learned Perceptual Image Patch Similarity(LPIPS)improvement of 68.13%in predicting the average quality of images,and the computational cost of the model is reduced by 30.36%compared with other models.In addition,the high model can also effectively predict the unknown incoming flow condition.
基金The National Natural Science Foundation of China(No.72001107,72271120)the Fundamental Research Funds for the Central Universities(No.NS2024047,NP2024106)the China Postdoctoral Science Foundation(No.2020T130297,2019M660119).
文摘The presence of circles in the network maximum flow problem increases the complexity of the preflow algorithm.This study proposes a novel two-stage preflow algorithm to address this issue.First,this study proves that at least one zero-flow arc must be present when the flow of the network reaches its maximum value.This result indicates that the maximum flow of the network will remain constant if a zero-flow arc within a circle is removed;therefore,the maximum flow of each network without circles can be calculated.The first stage involves identifying the zero-flow arc in the circle when the network flow reaches its maximum.The second stage aims to remove the zero-flow arc identified and modified in the first stage,thereby producing a new network without circles.The maximum flow of the original looped network can be obtained by solving the maximum flow of the newly generated acyclic network.Finally,an example is provided to demonstrate the validity and feasibility of this algorithm.This algorithm not only improves computational efficiency but also provides new perspectives and tools for solving similar network optimization problems.
基金Supported by the National Natural Science Foundation of China(No.62304022)。
文摘This research introduces a spectrum-based physics-informed neural network(SP-PINN)model to significantly improve the accuracy of calculation of two-phase flow parameters,surpassing existing methods that have limitations in global and continuous data sampling.SP-PINNs address the challenges of traditional methods in terms of continuous sampling by integrating the spectral analysis and pressure correction into the Navier-Stokes(N-S)equations,enhancing the predictive accuracy especially in critical regions like gas-phase boundaries and velocity peaks.The novel introduction of a pressure-correction module within SP-PINNs mitigates prediction errors,achieving a substantial reduction to 1‰compared with the conventional physics-informed neural network(PINN)approaches.Experimental applications validate the model’s ability to accurately and rapidly predict flow parameters with different sampling time intervals,with the computation time of predicting unsampled data less than 0.01 s.Such advancements signify a 100-fold improvement over traditional DNS calculations,underscoring the model’s potential in the real-time calculation and analysis of multiphase flow dynamics.
基金supported in part by National Natural Science Foundation of China under Grant 52007026.
文摘With the widespread implementation of distributed generation(DG)and the integration of soft open point(SOP)into the distribution network(DN),the latter is steadily transitioning into a flexible distribution network(FDN),the calculation of carbon flow distribution in FDN is more difficult.To this end,this study constructs a model for low-carbon optimal operations within the FDN on the basis of enhanced carbon emission flow(CEF).First,the carbon emission characteristics of FDNs are scrutinized and an improved method for calculating carbon flow within these networks is proposed.Subsequently,a model for optimizing low-carbon operations within FDNs is formulated based on the refined CEF,which merges the specificities of DG and intelligent SOP.Finally,this model is scrutinized using an upgraded IEEE 33-node distribution system,a comparative analysis of the cases reveals that when DG and SOP are operated in a coordinated manner in the FDN,with the cost of electricity generation was reduced by 40.63 percent and the cost of carbon emissions by 10.18 percent.The findings indicate that the judicious optimization of areas exhibiting higher carbon flow rates can effectively enhance the economic efficiency of DN operations and curtail the carbon emissions of the overall network.
基金supported by the National Natural Science Foundation of China (No. 52274048)Beijing Natural Science Foundation (No. 3222037)。
文摘With the rapid development of deep learning neural networks,new solutions have emerged for addressing fluid flow problems in porous media.Combining data-driven approaches with physical constraints has become a hot research direction,with physics-informed neural networks(PINNs) being the most popular hybrid model.PINNs have gained widespread attention in subsurface fluid flow simulations due to their low computational resource requirements,fast training speeds,strong generalization capabilities,and broad applicability.Despite success in homogeneous settings,standard PINNs face challenges in accurately calculating flux between irregular Eulerian cells with disparate properties and capturing global field influences on local cells.This limits their suitability for heterogeneous reservoirs and the irregular Eulerian grids frequently used in reservoir.To address these challenges,this study proposes a physics-informed graph neural network(PIGNN) model.The PIGNN model treats the entire field as a whole,integrating information from neighboring grids and physical laws into the solution for the target grid,thereby improving the accuracy of solving partial differential equations in heterogeneous and Eulerian irregular grids.The optimized model was applied to pressure field prediction in a spatially heterogeneous reservoir,achieving an average L_(2) error and R_(2) score of 6.710×10^(-4)and 0.998,respectively,which confirms the effectiveness of model.Compared to the conventional PINN model,the average L_(2) error was reduced by 76.93%,the average R_(2) score increased by 3.56%.Moreover,evaluating robustness,training the PIGNN model using only 54% and 76% of the original data yielded average relative L_(2) error reductions of 58.63% and 56.22%,respectively,compared to the PINN model.These results confirm the superior performance of this approach compared to PINN.
基金supported by the National Natural Science Foundation of China(No.52174043)the Beijing Natural Science Foundation(No.3242019)+1 种基金the CNPC Innovation Foundation(No.2022DQ02-0208)the State Key Laboratory of Deep Oil and Gas(No.SKLD0G2024-KFZD-06).
文摘Two actual rocks drilled from a typical ultra-deep hydrocarbon reservoir in the Tarim Basin are selected to conduct in-situ stress-loading micro-focus CT scanning experiments.The gray images of rock microstructure at different stress loading stages are obtained.The U-Net fully convolutional neural network is utilized to achieve fine semantic segmentation of rock skeleton,pore space,and microfractures based on CT slice images of deep rocks.The three-dimensional digital rock models of deformed multiscale fractured-porous media at different stress loading stages are thereafter reconstructed,and the equivalent fracture-pore network models are finally extracted to explore the underlying mechanisms of gas-water two-phase flow at the pore-scale.Results indicate that,in the process of insitu stress loading,both the deep rocks have experienced three stages:linear elastic deformation,nonlinear plastic deformation,and shear failure.The micro-mechanical behavior greatly affects the dynamic deformation of rock microstructure and gas-water two-phase flow.In the linear elastic deformation stage,with the increase in in-situ stress,both the deep rocks are gradually compacted,leading to decreases in average pore radius,pore throat ratio,tortuosity,and water-phase relative permeability,while the coordination number nearly remains unchanged.In the plastic deformation stage,the synergistic influence of rock compaction and existence of micro-fractures typically exert a great effect on pore-throat topological properties and gas-water relative permeability.In the shear failure stage,due to the generation and propagation of micro-fractures inside the deep rock,the topological connectivity becomes better,fluid flow paths increase,and flow conductivity is promoted,thus leading to sharp increases in average pore radius and coordination number,rapid decreases in pore throat ratio and tortuosity,as well as remarkable improvement in relative permeability of gas phase and waterphase.
基金Supported by the by Key R&D Program of Gansu Province(No.23YFGA0063)the Key Talent Project of Gansu Province(No.2024RCXM57,2024RCXM22)the Major Science and Technology Special Program of Gansu Province(No.25ZYJA037).
文摘Precise traffic flow forecasting is essential for mitigating urban traffic congestion.However,it is difficult for existing methods to adequately capture the dynamic spatio-temporal characteristics and multiscale temporal dependencies of traffic flow.A traffic flow prediction model with multiscale temporal awareness and graph diffusion attention networks(MT-GDAN)is proposed to address these issues.Specifically,a graph diffusion attention module is constructed,which dynamically adjusts and calculates the weights of neighboring nodes in the graph structure using a random graph attention network(GAT)and captures the spatial characteristics of hidden nodes through an adaptive adjacency matrix,thus better exploiting the dynamic spatio-temporal properties of traffic flow.Secondly,a multiscale isometric convolutional network and bi-level routing attention are used to construct a multiscale temporal awareness module.The former extracts local information of traffic flow segments by convolution with different sizes of convolution kernels and then introduces isometric convolution to obtain the global temporal relationship between local features of traffic flow segments;the latter filters irrelevant spatio-temporal features at a coarse regional level and focuses locally on key points to more accurately capture the multiscale temporal dependencies of traffic flows.Experimental results reveal that the MT-GDAN model surpasses the mainstream baseline model in terms of forecasting accuracy and exhibits good prediction performance.
基金Sponsored by National Key Research and Development Program of China(Grant No.2020YEB1600500).
文摘Short⁃term traffic flow prediction plays a crucial role in the planning of intelligent transportation systems.Nowadays,there is a large amount of traffic flow data generated from the monitoring devices of urban road networks,which contains road network traffic information with high application value.In this study,an improved spatio⁃temporal attention transformer model(ISTA⁃transformer model)is proposed to provide a more accurate method for predicting multi⁃step short⁃term traffic flow based on monitoring data.By embedding a temporal attention layer and a spatial attention layer in the model,the model learns the relationship between traffic flows at different time intervals and different geographic locations,and realizes more accurate multi⁃step short⁃time flow prediction.Finally,we validate the superiority of the model with monitoring data spanning 15 days from 620 monitoring points in Qingdao,China.In the four time steps of prediction,the MAPE(Mean Absolute Percentage Error)values of ISTA⁃transformers prediction results are 0.22,0.29,0.37,and 0.38,respectively,and its prediction accuracy is usually better than that of six baseline models(Transformer,GRU,CNN,LSTM,Seq2Seq and LightGBM),which indicates that the proposed model in this paper always has a better ability to explain the prediction results with the time steps in the multi⁃step prediction.
基金supported by the Major Project of the National Social Science Fund of China,titled“Design Path Selection for the Mechanism of New and Old Growth Driver Conversion”(Grant No.18ZDA077)by the Joint Special Major Research Project of the Yangtze River Delta Economics and Social Development Research Center at Nanjing University and the Collaborative Innovation Center for China Economy(CICCE),titled“Practicing Innovation in China’s Development Economics for the Yangtze River Delta:From Industrial Clusters to Technological Clusters”(Grant No.CYD2022006).
文摘The global clustering of inventive talent shapes innovation capacity and drives economic growth.For China,this process is especially crucial in sustaining its development momentum.This paper draws on data from the EPO Worldwide Patent Statistical Database(PATSTAT)to extract global inventive talent mobility information and analyzes the spatial structural evolution of the global inventive talent flow network.The study finds that this network is undergoing a multi-polar transformation,characterized by the rising importance of a few central countries-such as the United States,Germany,and China-and the increasing marginalization of many peripheral countries.In response to this typical phenomenon,the paper constructs an endogenous migration model and conducts empirical testing using the Temporal Exponential Random Graph Model(TERGM).The results reveal several endogenous mechanisms driving global inventive talent flows,including reciprocity,path dependence,convergence effects,transitivity,and cyclic structures,all of which contribute to the network’s multi-polar trend.In addition,differences in regional industrial structures significantly influence talent mobility choices and are a decisive factor in the formation of poles within the multi-polar landscape.Based on these findings,it is suggested that efforts be made to foster two-way channels for talent exchange between China and other global innovation hubs,in order to enhance international collaboration and knowledge flow.We should aim to reduce the migration costs and institutional barriers faced by R&D personnel,thereby encouraging greater mobility of high-skilled talent.Furthermore,the government is advised to strategically leverage regional strengths in high-tech industries as a lever to capture competitive advantages in emerging technologies and products,ultimately strengthening the country’s position in the global innovation landscape.
基金supported by the National Natural Science Foundation of China(Grant Nos.42272338 and 41827807)Department of Transportation of Zhejiang Province,China(Grant No.202213).
文摘Physics-informed neural networks(PINNs)have prevailed as differentiable simulators to investigate flow in porous media.Despite recent progress PINNs have achieved,practical geotechnical scenarios cannot be readily simulated because conventional PINNs fail in discontinuous heterogeneous porous media or multi-layer strata when labeled data are missing.This work aims to develop a universal network structure to encode the mass continuity equation and Darcy’s law without labeled data.The finite element approximation,which can decompose a complex heterogeneous domain into simpler ones,is adopted to build the differentiable network.Without conventional DNNs,physics-encoded finite element network(PEFEN)can avoid spectral bias and learn high-frequency functions with sharp/steep gradients.PEFEN rigorously encodes Dirichlet and Neumann boundary conditions without training.Benefiting from its discretized formulation,the discontinuous heterogeneous hydraulic conductivity is readily embedded into the network.Three typical cases are reproduced to corroborate PEFEN’s superior performance over conventional PINNs and the PINN with mixed formulation.PEFEN is sparse and demonstrated to be capable of dealing with heterogeneity with much fewer training iterations(less than 1/30)than the improved PINN with mixed formulation.Thus,PEFEN saves energy and contributes to low-carbon AI for science.The last two cases focus on common geotechnical settings of impermeable sheet pile in singlelayer and multi-layer strata.PEFEN solves these cases with high accuracy,circumventing costly labeled data,extra computational burden,and additional treatment.Thus,this study warrants the further development and application of PEFEN as a novel differentiable network in porous flow of practical geotechnical engineering.
文摘Characteristics of the Internet traffic data flow are studied based on the chaos theory. A phase space that is isometric with the network dynamic system is reconstructed by using the single variable time series of a network flow. Some parameters, such as the correlative dimension and the Lyapunov exponent are calculated, and the chaos characteristic is proved to exist in Internet traffic data flows. A neural network model is construct- ed based on radial basis function （RBF） to forecast actual Internet traffic data flow. Simulation results show that, compared with other forecasts of the forward-feedback neural network, the forecast of the RBF neural network based on the chaos theory has faster learning capacity and higher forecasting accuracy.
文摘The explicit rate flow control mechanisms for ABR service are used to sharethe available bandwidth of a bottleneck link fairly and reasonably among many competitive users andto maintain the buffer queue length of a bottleneck switch connected to the link at a desired levelin order to avoid and control congestion in ATM networks. However, designing effective flow controlmechanisms for the service is known to be difficult because of the variety of dynamic parametersinvolved such as available link bandwidth, burst of the traffic, the distances between ABR sourcesand switches. In this paper, we present a fuzzy explicit rate flow control mechanism for ABRservice. The mechanism has a simple structure and is robust in the sense that the mechanism'sstability is not sensitive to the change in the number of active virtual connections (VCs). Manysimulations show that this mechanism can not only effectively avoid network congestion, but alsoensure fair share of the bandwidth for all active VCs regardless of the number of hops theytraverse. Additionally, it has the advantages of fast convergence, low oscillation, and high linkbandwidth utilization.
文摘Blockage is a kind of phenomenon frequently occurred in a transport network, in which the human beings are the moving subjects. The minimum flow of a network defined in this paper means the maximum flow quantity through the network in the seriously blocked situation. It is an important parameter in designing and operating a transport network, especially in an emergency evacuation network. A branch and bound method is presented to solve the minimum flow problem on the basis of the blocking flow theory and the algorithm and its application are illustrated by examples.
文摘Blockage is a kind of phenomenon occurring frequently in modern transportation network. This paper deals with the research work on the blocking now in a network with the help of network flow theory. The blockage phenomena can be divided intO local blockage and network blockage. In this paper, which deals mainly with the latter, the fundamental concepts and definitions of network blocking flow, blocking outset are presented and the related theorems are proved. It is proved that the sufficient and necessary condition for the emergence of a blocking now in a network is the existence of the blocking outset. The necessary conditions for the existence of the blocking outset in a network are analysed and the characteristic cutset of blockage which reflects the all possible situation of blocking nows in the network is defined.In the last part of the paper the mathematical model of the minimum blocking now is developed and the solution to a small network is given.
基金supported by NIH(R01 NS 35069)the Morris K.Udall Center of Excellence for Parkinson's Disease Research (NIH P50 NS 071675) at The Feinstein Institute for Medical Research,USA
文摘Neuroimaging of cerebral glucose metabolism and blood flow is ideally suited to assay widely-distributed brain circuits as a result of local molecular events and behavioral modulation in the central nervous system. With the progress in novel analytical methodology, this endeavor has succeeded in unraveling the mechanisms underlying a wide spectrum of neurodegenerative diseases. In particular, statistical brain mapping studies have made significant strides in describing the pathophysiology of Parkinson＇s disease （PD） and related disorders by providing signature biomarkers to determine the systemic abnormalities in brain function and evaluate disease progression, therapeutic responses, and clinical correlates in patients. In this article, we review the relevant clinical applications in patients in relation to healthy volunteers with a focus on the generation of unique spatial covariance patterns associated with the motor and cognitive symptoms underlying PD. These characteristic biomarkers can be potentially used not only to improve patient recruitment but also to predict outcomes in clinical trials.
基金National Natural Science Foundation of China under Grant No. 60903196,60903175National Critical Patented Projects in the Next Generation Broadband Wireless Mobile Communication Network under Grant No. 2010ZX03006-001-01+1 种基金National High Technical Research and Development Program of China under Grant No. 2009AA01Z418Educational Commission of Hubei Province of China under Grant No. D20114401
文摘Although the wireless network is widely used in many fields,its characteristics such as high bit error rate and broadcast links may block its development.Network coding is an artistic way to exploit its intrinsic characteristics to increase the network reliability.Some people research network coding schemes for inter-flow or intra-flow,each type with its own advantages and disadvantages.In this paper,we propose a new mechanism,called MM-NCOPE,which integrates the idea of inter-flow and intra-flow coding.On the one hand,MM-NCOPE utilizes random liner coding to encode the NCOPE packets while NCOPE is a sub-protocol for optimizing the COPE algorithm by iteration.In NCOPE,packets are automatically matched by size to be coded.As a result,it improves the coding gain in some level.On the other hand,we adopt the partial Acknowledgement retransmission scheme to achieve high compactness and robustness.ACK is an independent packet with the highest priority rather than a part of the data packets.Compared with existing works on opportunistic network coding,our approach ensures the reliability of wireless links and improves the coding gain.
