With the advancement of Vehicle-to-Everything(V2X)technology,efficient resource allocation in dynamic vehicular networks has become a critical challenge for achieving optimal performance.Existing methods suffer from h...With the advancement of Vehicle-to-Everything(V2X)technology,efficient resource allocation in dynamic vehicular networks has become a critical challenge for achieving optimal performance.Existing methods suffer from high computational complexity and decision latency under high-density traffic and heterogeneous network conditions.To address these challenges,this study presents an innovative framework that combines Graph Neural Networks(GNNs)with a Double Deep Q-Network(DDQN),utilizing dynamic graph structures and reinforcement learning.An adaptive neighbor sampling mechanism is introduced to dynamically select the most relevant neighbors based on interference levels and network topology,thereby improving decision accuracy and efficiency.Meanwhile,the framework models communication links as nodes and interference relationships as edges,effectively capturing the direct impact of interference on resource allocation while reducing computational complexity and preserving critical interaction information.Employing an aggregation mechanism based on the Graph Attention Network(GAT),it dynamically adjusts the neighbor sampling scope and performs attention-weighted aggregation based on node importance,ensuring more efficient and adaptive resource management.This design ensures reliable Vehicle-to-Vehicle(V2V)communication while maintaining high Vehicle-to-Infrastructure(V2I)throughput.The framework retains the global feature learning capabilities of GNNs and supports distributed network deployment,allowing vehicles to extract low-dimensional graph embeddings from local observations for real-time resource decisions.Experimental results demonstrate that the proposed method significantly reduces computational overhead,mitigates latency,and improves resource utilization efficiency in vehicular networks under complex traffic scenarios.This research not only provides a novel solution to resource allocation challenges in V2X networks but also advances the application of DDQN in intelligent transportation systems,offering substantial theoretical significance and practical value.展开更多
The electric double layer(EDL),formed by charge adsorption at the electrolyte–electrode interface,constitutes the microenvironment governing electrochemical reactions.However,due to scale mismatch between the EDL thi...The electric double layer(EDL),formed by charge adsorption at the electrolyte–electrode interface,constitutes the microenvironment governing electrochemical reactions.However,due to scale mismatch between the EDL thickness and electrode topography,solving the two-dimensional(2D)nonhomogeneous Poisson–Nernst–Planck(N-PNP)equations remains computationally intractable.This limitation hinders understanding of fundamental phenomena such as curvature-driven instabilities in 2D EDL.Here,we propose a dimensionality-decomposition strategy embedding a fully connected neural network(FCNN)to solve 2D N-PNP equations,in which the FCNN is trained on key electrochemical parameters by reducing the electrostatic boundary into multiple equivalent 1D representations.Through a representative case of LiPF6 reduction on lithium metal half-cell,nucleus size is unexpectedly found to have an important influence on dendrite morphology and tip kinetics.This work paves the way for bridging nanoscale and macroscale simulations with expandability to 2D situations of other 1D EDL models.展开更多
With the rapid development ofmobile Internet,spatial crowdsourcing has becomemore andmore popular.Spatial crowdsourcing consists of many different types of applications,such as spatial crowd-sensing services.In terms ...With the rapid development ofmobile Internet,spatial crowdsourcing has becomemore andmore popular.Spatial crowdsourcing consists of many different types of applications,such as spatial crowd-sensing services.In terms of spatial crowd-sensing,it collects and analyzes traffic sensing data from clients like vehicles and traffic lights to construct intelligent traffic prediction models.Besides collecting sensing data,spatial crowdsourcing also includes spatial delivery services like DiDi and Uber.Appropriate task assignment and worker selection dominate the service quality for spatial crowdsourcing applications.Previous research conducted task assignments via traditional matching approaches or using simple network models.However,advanced mining methods are lacking to explore the relationship between workers,task publishers,and the spatio-temporal attributes in tasks.Therefore,in this paper,we propose a Deep Double Dueling Spatial-temporal Q Network(D3SQN)to adaptively learn the spatialtemporal relationship between task,task publishers,and workers in a dynamic environment to achieve optimal allocation.Specifically,D3SQNis revised through reinforcement learning by adding a spatial-temporal transformer that can estimate the expected state values and action advantages so as to improve the accuracy of task assignments.Extensive experiments are conducted over real data collected fromDiDi and ELM,and the simulation results verify the effectiveness of our proposed models.展开更多
Double Q-learning has been shown to be effective in reinforcement learning scenarios when the reward system is stochastic. We apply the idea of double learning that this algorithm uses to Sarsa and Expected Sarsa, pro...Double Q-learning has been shown to be effective in reinforcement learning scenarios when the reward system is stochastic. We apply the idea of double learning that this algorithm uses to Sarsa and Expected Sarsa, producing two new algorithms called Double Sarsa and Double Expected Sarsa that are shown to be more robust than their single counterparts when rewards are stochastic. We find that these algorithms add a significant amount of stability in the learning process at only a minor computational cost, which leads to higher returns when using an on-policy algorithm. We then use shallow and deep neural networks to approximate the actionvalue, and show that Double Sarsa and Double Expected Sarsa are much more stable after convergence and can collect larger rewards than the single versions.展开更多
Hydrodynamic deep drawing assisted by radial pressure is an advanced sheet forming technology with great advantages such as higher drawing ratio, good surface quality and higher dimensional accuracy. In this process, ...Hydrodynamic deep drawing assisted by radial pressure is an advanced sheet forming technology with great advantages such as higher drawing ratio, good surface quality and higher dimensional accuracy. In this process, both the bottom surface and the peripheral edge of sheets are under hydrodynamic pressure, so that the forming procedure is more uniform with low failure probability. Multi-layered sheets with complex geometries could be formed more easily with this technique compared with other traditional methods. Rupture is the main irrecoverable failure form in sheet forming processes. Prediction of rupture occurrence is of great importance for determining and optimizing the proper process parameters. In this research, a theoretical model was proposed to calculate the critical rupture pressure in production of double layered conical parts with hydrodynamic deep drawing process assisted by radial pressure. The effects of other process parameters on critical rupture pressure, such as punch tip radius, drawing ratio, coefficient of friction, sheet thickness and material properties were also discussed. The proposed model was compared with finite element simulation and validated by experiments on Al1050/St13 double layered sheets, where a good agreement was found with analytical results.展开更多
BACKGROUND Multiple linear stapler firings during double stapling technique(DST)after laparoscopic low anterior resection(LAR)are associated with an increased risk of anastomotic leakage(AL).However,it is difficult to...BACKGROUND Multiple linear stapler firings during double stapling technique(DST)after laparoscopic low anterior resection(LAR)are associated with an increased risk of anastomotic leakage(AL).However,it is difficult to predict preoperatively the need for multiple linear stapler cartridges during DST anastomosis.AIM To develop a deep learning model to predict multiple firings during DST anastomosis based on pelvic magnetic resonance imaging(MRI).METHODS We collected 9476 MR images from 328 mid-low rectal cancer patients undergoing LAR with DST anastomosis,which were randomly divided into a training set(n=260)and testing set(n=68).Binary logistic regression was adopted to create a clinical model using six factors.The sequence of fast spin-echo T2-weighted MRI of the entire pelvis was segmented and analyzed.Pure-image and clinical-image integrated deep learning models were constructed using the mask region-based convolutional neural network segmentation tool and three-dimensional convolutional networks.Sensitivity,specificity,accuracy,positive predictive value(PPV),and area under the receiver operating characteristic curve(AUC)was calculated for each model.RESULTS The prevalence of≥3 linear stapler cartridges was 17.7%(58/328).The prevalence of AL was statistically significantly higher in patients with≥3 cartridges compared to those with≤2 cartridges(25.0%vs 11.8%,P=0.018).Preoperative carcinoembryonic antigen level>5 ng/mL(OR=2.11,95%CI 1.08-4.12,P=0.028)and tumor size≥5 cm(OR=3.57,95%CI 1.61-7.89,P=0.002)were recognized as independent risk factors for use of≥3 linear stapler cartridges.Diagnostic performance was better with the integrated model(accuracy=94.1%,PPV=87.5%,and AUC=0.88)compared with the clinical model(accuracy=86.7%,PPV=38.9%,and AUC=0.72)and the image model(accuracy=91.2%,PPV=83.3%,and AUC=0.81).CONCLUSION MRI-based deep learning model can predict the use of≥3 linear stapler cartridges during DST anastomosis in laparoscopic LAR surgery.This model might help determine the best anastomosis strategy by avoiding DST when there is a high probability of the need for≥3 linear stapler cartridges.展开更多
The problem of cavity stability widely exists in deep underground engineering and energy exploitation.First,the stress field of the surrounding rock under the uniform stress field is deduced based on a post-peak stren...The problem of cavity stability widely exists in deep underground engineering and energy exploitation.First,the stress field of the surrounding rock under the uniform stress field is deduced based on a post-peak strength drop model considering the rock’s characteristics of constant modulus and double moduli.Then,the orthogonal non-associative flow rule is used to establish the displacement of the surrounding rock under constant modulus and double moduli,respectively,considering the stiffness degradation and dilatancy effects in the plastic region and assuming that the elastic strain in the plastic region satisfies the elastic constitutive relationship.Finally,the evolution of the displacement in the surrounding rock is analyzed under the effects of the double modulus characteristics,the strength drop,the stiffness degradation,and the dilatancy.The results show that the displacement solutions of the surrounding rock under constant modulus and double moduli have a unified expression.The coefficients of the expression are related to the stress field of the original rock,the elastic constant of the surrounding rock,the strength parameters,and the dilatancy angle.The strength drop,the stiffness degradation,and the dilatancy effects all have effects on the displacement.The effects can be characterized by quantitative relationships.展开更多
A deep trench super-junction LDMOS with double charge compensation layer(DC DT SJ LDMOS)is proposed in this paper.Due to the capacitance effect of the deep trench which is known as silicon-insulator-silicon(SIS)capaci...A deep trench super-junction LDMOS with double charge compensation layer(DC DT SJ LDMOS)is proposed in this paper.Due to the capacitance effect of the deep trench which is known as silicon-insulator-silicon(SIS)capacitance,the charge balance in the super-junction region of the conventional deep trench SJ LDMOS(Con.DT SJ LDMOS)device will be broken,resulting in breakdown voltage(BV)of the device drops.DC DT SJ LDMOS solves the SIS capacitance effect by adding a vertical variable doped charge compensation layer and a triangular charge compensation layer inside the Con.DT SJ LDMOS device.Therefore,the drift region reaches an ideal charge balance state again.The electric field is optimized by double charge compensation and gate field plate so that the breakdown voltage of the proposed device is improved sharply,meanwhile the enlarged on-current region reduces its specific on-resistance.The simulation results show that compared with the Con.DT SJ LD-MOS,the BV of the DC DT SJ LDMOS has been increased from 549.5 to 705.5 V,and the R_(on,sp) decreased to 23.7 mΩ·cm^(2).展开更多
Under deep and complex geological conditions,severe deformation occurs at intersection points of Y-type roadways with large cross sections during engineering projects in coal mines,especially at junction arches.Based ...Under deep and complex geological conditions,severe deformation occurs at intersection points of Y-type roadways with large cross sections during engineering projects in coal mines,especially at junction arches.Based on in-situ investigations and theoretical studies,we have summarized typical forms of destruction and identified high stress and unrestricted support at both sides of junction arch as its main causes.In this study,we also presented double-directional control bolt support technology for a large Y-type span intersection,applied to deep intersection engineering in the Jiahe Coal Mine,which has proved effective.展开更多
In response to the limitations and low computational efficiency of one-dimensional water and sediment models in representing complex hydrological conditions, this paper proposes a dual branch convolution method based ...In response to the limitations and low computational efficiency of one-dimensional water and sediment models in representing complex hydrological conditions, this paper proposes a dual branch convolution method based on deep learning. This method utilizes the ability of deep learning to extract data features and introduces a dual branch convolutional network to handle the non-stationary and nonlinear characteristics of noise and reservoir sediment transport data. This method combines permutation variant structure to preserve the original time series information, constructs a corresponding time series model, models and analyzes the changes in the outbound water and sediment sequence, and can more accurately predict the future trend of outbound sediment changes based on the current sequence changes. The experimental results show that the DCON model established in this paper has good predictive performance in monthly, bimonthly, seasonal, and semi-annual predictions, with determination coefficients of 0.891, 0.898, 0.921, and 0.931, respectively. The results can provide more reference schemes for personnel formulating reservoir scheduling plans. Although this study has shown good applicability in predicting sediment discharge, it has not been able to make timely predictions for some non-periodic events in reservoirs. Therefore, future research will gradually incorporate monitoring devices to obtain more comprehensive data, in order to further validate and expand the conclusions of this study.展开更多
By integrating deep neural networks with reinforcement learning,the Double Deep Q Network(DDQN)algorithm overcomes the limitations of Q-learning in handling continuous spaces and is widely applied in the path planning...By integrating deep neural networks with reinforcement learning,the Double Deep Q Network(DDQN)algorithm overcomes the limitations of Q-learning in handling continuous spaces and is widely applied in the path planning of mobile robots.However,the traditional DDQN algorithm suffers from sparse rewards and inefficient utilization of high-quality data.Targeting those problems,an improved DDQN algorithm based on average Q-value estimation and reward redistribution was proposed.First,to enhance the precision of the target Q-value,the average of multiple previously learned Q-values from the target Q network is used to replace the single Q-value from the current target Q network.Next,a reward redistribution mechanism is designed to overcome the sparse reward problem by adjusting the final reward of each action using the round reward from trajectory information.Additionally,a reward-prioritized experience selection method is introduced,which ranks experience samples according to reward values to ensure frequent utilization of high-quality data.Finally,simulation experiments are conducted to verify the effectiveness of the proposed algorithm in fixed-position scenario and random environments.The experimental results show that compared to the traditional DDQN algorithm,the proposed algorithm achieves shorter average running time,higher average return and fewer average steps.The performance of the proposed algorithm is improved by 11.43%in the fixed scenario and 8.33%in random environments.It not only plans economic and safe paths but also significantly improves efficiency and generalization in path planning,making it suitable for widespread application in autonomous navigation and industrial automation.展开更多
At present,energy consumption is one of the main bottlenecks in autonomous mobile robot development.To address the challenge of high energy consumption in path planning for autonomous mobile robots navigating unknown ...At present,energy consumption is one of the main bottlenecks in autonomous mobile robot development.To address the challenge of high energy consumption in path planning for autonomous mobile robots navigating unknown and complex environments,this paper proposes an Attention-Enhanced Dueling Deep Q-Network(ADDueling DQN),which integrates a multi-head attention mechanism and a prioritized experience replay strategy into a Dueling-DQN reinforcement learning framework.A multi-objective reward function,centered on energy efficiency,is designed to comprehensively consider path length,terrain slope,motion smoothness,and obstacle avoidance,enabling optimal low-energy trajectory generation in 3D space from the source.The incorporation of a multihead attention mechanism allows the model to dynamically focus on energy-critical state features—such as slope gradients and obstacle density—thereby significantly improving its ability to recognize and avoid energy-intensive paths.Additionally,the prioritized experience replay mechanism accelerates learning from key decision-making experiences,suppressing inefficient exploration and guiding the policy toward low-energy solutions more rapidly.The effectiveness of the proposed path planning algorithm is validated through simulation experiments conducted in multiple off-road scenarios.Results demonstrate that AD-Dueling DQN consistently achieves the lowest average energy consumption across all tested environments.Moreover,the proposed method exhibits faster convergence and greater training stability compared to baseline algorithms,highlighting its global optimization capability under energy-aware objectives in complex terrains.This study offers an efficient and scalable intelligent control strategy for the development of energy-conscious autonomous navigation systems.展开更多
In Chinese modernization and social development level enhances unceasingly, and under the background of deepening the process of urbanization, urban development level in China has been an unprecedented increase, espec...In Chinese modernization and social development level enhances unceasingly, and under the background of deepening the process of urbanization, urban development level in China has been an unprecedented increase, especially with the constant development of information technology, make our country construction technology has been constantly strengthened, all kinds of tunnel construction, underground construction, high-rise buildings appear constantly, higher and more strict requirements are put forward for deep foundation pit engineering in terms of quantity and construction quality. In this paper, a detailed analysis is carried out on the simulation and optimization of the double-row pile supporting structure of deep foundation pit, which lays a solid foundation for the further improvement of the modern construction technology level in China.展开更多
Deep stochastic configuration networks(DSCNs)produce redundant hidden nodes and connections during training,which complicates their model structures.Aiming at the above problems,this paper proposes a double pruning st...Deep stochastic configuration networks(DSCNs)produce redundant hidden nodes and connections during training,which complicates their model structures.Aiming at the above problems,this paper proposes a double pruning structure design algorithm for DSCNs based on mutual information and relevance.During the training process,the mutual information algorithm is used to calculate and sort the importance scores of the nodes in each hidden layer in a layer-by-layer manner,the node pruning rate of each layer is set according to the depth of the DSCN at the current time,the nodes that contribute little to the model are deleted,and the network-related parameters are updated.When the model completes the configuration procedure,the correlation evaluation strategy is used to sort the global connection weights and delete insignificance connections;then,the network parameters are updated after pruning is completed.The experimental results show that the proposed structure design method can effectively compress the scale of a DSCN model and improve its modeling speed;the model accuracy loss is small,and fine-tuning for accuracy restoration is not needed.The obtained DSCN model has certain application value in the field of regression analysis.展开更多
The multi-agent path planning problem presents significant challenges in dynamic environments,primarily due to the ever-changing positions of obstacles and the complex interactions between agents’actions.These factor...The multi-agent path planning problem presents significant challenges in dynamic environments,primarily due to the ever-changing positions of obstacles and the complex interactions between agents’actions.These factors contribute to a tendency for the solution to converge slowly,and in some cases,diverge altogether.In addressing this issue,this paper introduces a novel approach utilizing a double dueling deep Q-network(D3QN),tailored for dynamic multi-agent environments.A novel reward function based on multi-agent positional constraints is designed,and a training strategy based on incremental learning is performed to achieve collaborative path planning of multiple agents.Moreover,the greedy and Boltzmann probability selection policy is introduced for action selection and avoiding convergence to local extremum.To match radar and image sensors,a convolutional neural network-long short-term memory(CNN-LSTM)architecture is constructed to extract the feature of multi-source measurement as the input of the D3QN.The algorithm’s efficacy and reliability are validated in a simulated environment,utilizing robot operating system and Gazebo.The simulation results show that the proposed algorithm provides a real-time solution for path planning tasks in dynamic scenarios.In terms of the average success rate and accuracy,the proposed method is superior to other deep learning algorithms,and the convergence speed is also improved.展开更多
AIM: To assess “top-down” treatment for deep remission of early moderate to severe Crohn’s disease (CD) by double balloon enteroscopy.
The recent surge of mobile subscribers and user data traffic has accelerated the telecommunication sector towards the adoption of the fifth-generation (5G) mobile networks. Cloud radio access network (CRAN) is a promi...The recent surge of mobile subscribers and user data traffic has accelerated the telecommunication sector towards the adoption of the fifth-generation (5G) mobile networks. Cloud radio access network (CRAN) is a prominent framework in the 5G mobile network to meet the above requirements by deploying low-cost and intelligent multiple distributed antennas known as remote radio heads (RRHs). However, achieving the optimal resource allocation (RA) in CRAN using the traditional approach is still challenging due to the complex structure. In this paper, we introduce the convolutional neural network-based deep Q-network (CNN-DQN) to balance the energy consumption and guarantee the user quality of service (QoS) demand in downlink CRAN. We first formulate the Markov decision process (MDP) for energy efficiency (EE) and build up a 3-layer CNN to capture the environment feature as an input state space. We then use DQN to turn on/off the RRHs dynamically based on the user QoS demand and energy consumption in the CRAN. Finally, we solve the RA problem based on the user constraint and transmit power to guarantee the user QoS demand and maximize the EE with a minimum number of active RRHs. In the end, we conduct the simulation to compare our proposed scheme with nature DQN and the traditional approach.展开更多
To optimize machine allocation and task dispatching in smart manufacturing factories, this paper proposes a manufacturing resource scheduling framework based on reinforcement learning(RL). The framework formulates the...To optimize machine allocation and task dispatching in smart manufacturing factories, this paper proposes a manufacturing resource scheduling framework based on reinforcement learning(RL). The framework formulates the entire scheduling process as a multi-stage sequential decision problem, and further obtains the scheduling order by the combination of deep convolutional neural network(CNN) and improved deep Q-network(DQN). Specifically, with respect to the representation of the Markov decision process(MDP), the feature matrix is considered as the state space and a set of heuristic dispatching rules are denoted as the action space. In addition, the deep CNN is employed to approximate the state-action values, and the double dueling deep Qnetwork with prioritized experience replay and noisy network(D3QPN2) is adopted to determine the appropriate action according to the current state. In the experiments, compared with the traditional heuristic method, the proposed method is able to learn high-quality scheduling policy and achieve shorter makespan on the standard public datasets.展开更多
基金Project ZR2023MF111 supported by Shandong Provincial Natural Science Foundation。
文摘With the advancement of Vehicle-to-Everything(V2X)technology,efficient resource allocation in dynamic vehicular networks has become a critical challenge for achieving optimal performance.Existing methods suffer from high computational complexity and decision latency under high-density traffic and heterogeneous network conditions.To address these challenges,this study presents an innovative framework that combines Graph Neural Networks(GNNs)with a Double Deep Q-Network(DDQN),utilizing dynamic graph structures and reinforcement learning.An adaptive neighbor sampling mechanism is introduced to dynamically select the most relevant neighbors based on interference levels and network topology,thereby improving decision accuracy and efficiency.Meanwhile,the framework models communication links as nodes and interference relationships as edges,effectively capturing the direct impact of interference on resource allocation while reducing computational complexity and preserving critical interaction information.Employing an aggregation mechanism based on the Graph Attention Network(GAT),it dynamically adjusts the neighbor sampling scope and performs attention-weighted aggregation based on node importance,ensuring more efficient and adaptive resource management.This design ensures reliable Vehicle-to-Vehicle(V2V)communication while maintaining high Vehicle-to-Infrastructure(V2I)throughput.The framework retains the global feature learning capabilities of GNNs and supports distributed network deployment,allowing vehicles to extract low-dimensional graph embeddings from local observations for real-time resource decisions.Experimental results demonstrate that the proposed method significantly reduces computational overhead,mitigates latency,and improves resource utilization efficiency in vehicular networks under complex traffic scenarios.This research not only provides a novel solution to resource allocation challenges in V2X networks but also advances the application of DDQN in intelligent transportation systems,offering substantial theoretical significance and practical value.
基金supported by the National Natural Science Foundation of China(Grant Nos.92472207,52472223,and 92572301)。
文摘The electric double layer(EDL),formed by charge adsorption at the electrolyte–electrode interface,constitutes the microenvironment governing electrochemical reactions.However,due to scale mismatch between the EDL thickness and electrode topography,solving the two-dimensional(2D)nonhomogeneous Poisson–Nernst–Planck(N-PNP)equations remains computationally intractable.This limitation hinders understanding of fundamental phenomena such as curvature-driven instabilities in 2D EDL.Here,we propose a dimensionality-decomposition strategy embedding a fully connected neural network(FCNN)to solve 2D N-PNP equations,in which the FCNN is trained on key electrochemical parameters by reducing the electrostatic boundary into multiple equivalent 1D representations.Through a representative case of LiPF6 reduction on lithium metal half-cell,nucleus size is unexpectedly found to have an important influence on dendrite morphology and tip kinetics.This work paves the way for bridging nanoscale and macroscale simulations with expandability to 2D situations of other 1D EDL models.
基金supported in part by the Pioneer and Leading Goose R&D Program of Zhejiang Province under Grant 2022C01083 (Dr.Yu Li,https://zjnsf.kjt.zj.gov.cn/)Pioneer and Leading Goose R&D Program of Zhejiang Province under Grant 2023C01217 (Dr.Yu Li,https://zjnsf.kjt.zj.gov.cn/).
文摘With the rapid development ofmobile Internet,spatial crowdsourcing has becomemore andmore popular.Spatial crowdsourcing consists of many different types of applications,such as spatial crowd-sensing services.In terms of spatial crowd-sensing,it collects and analyzes traffic sensing data from clients like vehicles and traffic lights to construct intelligent traffic prediction models.Besides collecting sensing data,spatial crowdsourcing also includes spatial delivery services like DiDi and Uber.Appropriate task assignment and worker selection dominate the service quality for spatial crowdsourcing applications.Previous research conducted task assignments via traditional matching approaches or using simple network models.However,advanced mining methods are lacking to explore the relationship between workers,task publishers,and the spatio-temporal attributes in tasks.Therefore,in this paper,we propose a Deep Double Dueling Spatial-temporal Q Network(D3SQN)to adaptively learn the spatialtemporal relationship between task,task publishers,and workers in a dynamic environment to achieve optimal allocation.Specifically,D3SQNis revised through reinforcement learning by adding a spatial-temporal transformer that can estimate the expected state values and action advantages so as to improve the accuracy of task assignments.Extensive experiments are conducted over real data collected fromDiDi and ELM,and the simulation results verify the effectiveness of our proposed models.
文摘Double Q-learning has been shown to be effective in reinforcement learning scenarios when the reward system is stochastic. We apply the idea of double learning that this algorithm uses to Sarsa and Expected Sarsa, producing two new algorithms called Double Sarsa and Double Expected Sarsa that are shown to be more robust than their single counterparts when rewards are stochastic. We find that these algorithms add a significant amount of stability in the learning process at only a minor computational cost, which leads to higher returns when using an on-policy algorithm. We then use shallow and deep neural networks to approximate the actionvalue, and show that Double Sarsa and Double Expected Sarsa are much more stable after convergence and can collect larger rewards than the single versions.
文摘Hydrodynamic deep drawing assisted by radial pressure is an advanced sheet forming technology with great advantages such as higher drawing ratio, good surface quality and higher dimensional accuracy. In this process, both the bottom surface and the peripheral edge of sheets are under hydrodynamic pressure, so that the forming procedure is more uniform with low failure probability. Multi-layered sheets with complex geometries could be formed more easily with this technique compared with other traditional methods. Rupture is the main irrecoverable failure form in sheet forming processes. Prediction of rupture occurrence is of great importance for determining and optimizing the proper process parameters. In this research, a theoretical model was proposed to calculate the critical rupture pressure in production of double layered conical parts with hydrodynamic deep drawing process assisted by radial pressure. The effects of other process parameters on critical rupture pressure, such as punch tip radius, drawing ratio, coefficient of friction, sheet thickness and material properties were also discussed. The proposed model was compared with finite element simulation and validated by experiments on Al1050/St13 double layered sheets, where a good agreement was found with analytical results.
基金Shanghai Jiaotong University,No.YG2019QNB24This study was reviewed and approved by Ruijin Hospital Ethics Committee(Approval No.2019-82).
文摘BACKGROUND Multiple linear stapler firings during double stapling technique(DST)after laparoscopic low anterior resection(LAR)are associated with an increased risk of anastomotic leakage(AL).However,it is difficult to predict preoperatively the need for multiple linear stapler cartridges during DST anastomosis.AIM To develop a deep learning model to predict multiple firings during DST anastomosis based on pelvic magnetic resonance imaging(MRI).METHODS We collected 9476 MR images from 328 mid-low rectal cancer patients undergoing LAR with DST anastomosis,which were randomly divided into a training set(n=260)and testing set(n=68).Binary logistic regression was adopted to create a clinical model using six factors.The sequence of fast spin-echo T2-weighted MRI of the entire pelvis was segmented and analyzed.Pure-image and clinical-image integrated deep learning models were constructed using the mask region-based convolutional neural network segmentation tool and three-dimensional convolutional networks.Sensitivity,specificity,accuracy,positive predictive value(PPV),and area under the receiver operating characteristic curve(AUC)was calculated for each model.RESULTS The prevalence of≥3 linear stapler cartridges was 17.7%(58/328).The prevalence of AL was statistically significantly higher in patients with≥3 cartridges compared to those with≤2 cartridges(25.0%vs 11.8%,P=0.018).Preoperative carcinoembryonic antigen level>5 ng/mL(OR=2.11,95%CI 1.08-4.12,P=0.028)and tumor size≥5 cm(OR=3.57,95%CI 1.61-7.89,P=0.002)were recognized as independent risk factors for use of≥3 linear stapler cartridges.Diagnostic performance was better with the integrated model(accuracy=94.1%,PPV=87.5%,and AUC=0.88)compared with the clinical model(accuracy=86.7%,PPV=38.9%,and AUC=0.72)and the image model(accuracy=91.2%,PPV=83.3%,and AUC=0.81).CONCLUSION MRI-based deep learning model can predict the use of≥3 linear stapler cartridges during DST anastomosis in laparoscopic LAR surgery.This model might help determine the best anastomosis strategy by avoiding DST when there is a high probability of the need for≥3 linear stapler cartridges.
基金Project supported by the National Natural Science Foundation of China and Shandong Province Joint Program(No.U1806209)the National Natural Science Foundation of China(Nos.51774196 and 51774194)and Shandong University of Science and Technology(SDUST)Research Fund(No.2019TDJH101)。
文摘The problem of cavity stability widely exists in deep underground engineering and energy exploitation.First,the stress field of the surrounding rock under the uniform stress field is deduced based on a post-peak strength drop model considering the rock’s characteristics of constant modulus and double moduli.Then,the orthogonal non-associative flow rule is used to establish the displacement of the surrounding rock under constant modulus and double moduli,respectively,considering the stiffness degradation and dilatancy effects in the plastic region and assuming that the elastic strain in the plastic region satisfies the elastic constitutive relationship.Finally,the evolution of the displacement in the surrounding rock is analyzed under the effects of the double modulus characteristics,the strength drop,the stiffness degradation,and the dilatancy.The results show that the displacement solutions of the surrounding rock under constant modulus and double moduli have a unified expression.The coefficients of the expression are related to the stress field of the original rock,the elastic constant of the surrounding rock,the strength parameters,and the dilatancy angle.The strength drop,the stiffness degradation,and the dilatancy effects all have effects on the displacement.The effects can be characterized by quantitative relationships.
文摘A deep trench super-junction LDMOS with double charge compensation layer(DC DT SJ LDMOS)is proposed in this paper.Due to the capacitance effect of the deep trench which is known as silicon-insulator-silicon(SIS)capacitance,the charge balance in the super-junction region of the conventional deep trench SJ LDMOS(Con.DT SJ LDMOS)device will be broken,resulting in breakdown voltage(BV)of the device drops.DC DT SJ LDMOS solves the SIS capacitance effect by adding a vertical variable doped charge compensation layer and a triangular charge compensation layer inside the Con.DT SJ LDMOS device.Therefore,the drift region reaches an ideal charge balance state again.The electric field is optimized by double charge compensation and gate field plate so that the breakdown voltage of the proposed device is improved sharply,meanwhile the enlarged on-current region reduces its specific on-resistance.The simulation results show that compared with the Con.DT SJ LD-MOS,the BV of the DC DT SJ LDMOS has been increased from 549.5 to 705.5 V,and the R_(on,sp) decreased to 23.7 mΩ·cm^(2).
基金supported by the National Basic Research Program of China (No.2006CB202200)the Major Program of the National Natural Science Foundation of China (No.50490270)the Innovative Team Development Project of the Ministry of Education of China (No.IRT0656)
文摘Under deep and complex geological conditions,severe deformation occurs at intersection points of Y-type roadways with large cross sections during engineering projects in coal mines,especially at junction arches.Based on in-situ investigations and theoretical studies,we have summarized typical forms of destruction and identified high stress and unrestricted support at both sides of junction arch as its main causes.In this study,we also presented double-directional control bolt support technology for a large Y-type span intersection,applied to deep intersection engineering in the Jiahe Coal Mine,which has proved effective.
基金NationalNatural Science Foundation of China(U2243236,51879115,U2243215),Recipients of funds:Xinjie Li,URL:https://www.nsfc.gov.cn/(accessed on 25 November 2024).
文摘In response to the limitations and low computational efficiency of one-dimensional water and sediment models in representing complex hydrological conditions, this paper proposes a dual branch convolution method based on deep learning. This method utilizes the ability of deep learning to extract data features and introduces a dual branch convolutional network to handle the non-stationary and nonlinear characteristics of noise and reservoir sediment transport data. This method combines permutation variant structure to preserve the original time series information, constructs a corresponding time series model, models and analyzes the changes in the outbound water and sediment sequence, and can more accurately predict the future trend of outbound sediment changes based on the current sequence changes. The experimental results show that the DCON model established in this paper has good predictive performance in monthly, bimonthly, seasonal, and semi-annual predictions, with determination coefficients of 0.891, 0.898, 0.921, and 0.931, respectively. The results can provide more reference schemes for personnel formulating reservoir scheduling plans. Although this study has shown good applicability in predicting sediment discharge, it has not been able to make timely predictions for some non-periodic events in reservoirs. Therefore, future research will gradually incorporate monitoring devices to obtain more comprehensive data, in order to further validate and expand the conclusions of this study.
基金funded by National Natural Science Foundation of China(No.62063006)Guangxi Science and Technology Major Program(No.2022AA05002)+1 种基金Key Laboratory of AI and Information Processing(Hechi University),Education Department of Guangxi Zhuang Autonomous Region(No.2022GXZDSY003)Central Leading Local Science and Technology Development Fund Project of Wuzhou(No.202201001).
文摘By integrating deep neural networks with reinforcement learning,the Double Deep Q Network(DDQN)algorithm overcomes the limitations of Q-learning in handling continuous spaces and is widely applied in the path planning of mobile robots.However,the traditional DDQN algorithm suffers from sparse rewards and inefficient utilization of high-quality data.Targeting those problems,an improved DDQN algorithm based on average Q-value estimation and reward redistribution was proposed.First,to enhance the precision of the target Q-value,the average of multiple previously learned Q-values from the target Q network is used to replace the single Q-value from the current target Q network.Next,a reward redistribution mechanism is designed to overcome the sparse reward problem by adjusting the final reward of each action using the round reward from trajectory information.Additionally,a reward-prioritized experience selection method is introduced,which ranks experience samples according to reward values to ensure frequent utilization of high-quality data.Finally,simulation experiments are conducted to verify the effectiveness of the proposed algorithm in fixed-position scenario and random environments.The experimental results show that compared to the traditional DDQN algorithm,the proposed algorithm achieves shorter average running time,higher average return and fewer average steps.The performance of the proposed algorithm is improved by 11.43%in the fixed scenario and 8.33%in random environments.It not only plans economic and safe paths but also significantly improves efficiency and generalization in path planning,making it suitable for widespread application in autonomous navigation and industrial automation.
文摘At present,energy consumption is one of the main bottlenecks in autonomous mobile robot development.To address the challenge of high energy consumption in path planning for autonomous mobile robots navigating unknown and complex environments,this paper proposes an Attention-Enhanced Dueling Deep Q-Network(ADDueling DQN),which integrates a multi-head attention mechanism and a prioritized experience replay strategy into a Dueling-DQN reinforcement learning framework.A multi-objective reward function,centered on energy efficiency,is designed to comprehensively consider path length,terrain slope,motion smoothness,and obstacle avoidance,enabling optimal low-energy trajectory generation in 3D space from the source.The incorporation of a multihead attention mechanism allows the model to dynamically focus on energy-critical state features—such as slope gradients and obstacle density—thereby significantly improving its ability to recognize and avoid energy-intensive paths.Additionally,the prioritized experience replay mechanism accelerates learning from key decision-making experiences,suppressing inefficient exploration and guiding the policy toward low-energy solutions more rapidly.The effectiveness of the proposed path planning algorithm is validated through simulation experiments conducted in multiple off-road scenarios.Results demonstrate that AD-Dueling DQN consistently achieves the lowest average energy consumption across all tested environments.Moreover,the proposed method exhibits faster convergence and greater training stability compared to baseline algorithms,highlighting its global optimization capability under energy-aware objectives in complex terrains.This study offers an efficient and scalable intelligent control strategy for the development of energy-conscious autonomous navigation systems.
文摘In Chinese modernization and social development level enhances unceasingly, and under the background of deepening the process of urbanization, urban development level in China has been an unprecedented increase, especially with the constant development of information technology, make our country construction technology has been constantly strengthened, all kinds of tunnel construction, underground construction, high-rise buildings appear constantly, higher and more strict requirements are put forward for deep foundation pit engineering in terms of quantity and construction quality. In this paper, a detailed analysis is carried out on the simulation and optimization of the double-row pile supporting structure of deep foundation pit, which lays a solid foundation for the further improvement of the modern construction technology level in China.
基金supported by the National Natural Science Foundation of China(62073006)the Beijing Natural Science Foundation of China(4212032)
文摘Deep stochastic configuration networks(DSCNs)produce redundant hidden nodes and connections during training,which complicates their model structures.Aiming at the above problems,this paper proposes a double pruning structure design algorithm for DSCNs based on mutual information and relevance.During the training process,the mutual information algorithm is used to calculate and sort the importance scores of the nodes in each hidden layer in a layer-by-layer manner,the node pruning rate of each layer is set according to the depth of the DSCN at the current time,the nodes that contribute little to the model are deleted,and the network-related parameters are updated.When the model completes the configuration procedure,the correlation evaluation strategy is used to sort the global connection weights and delete insignificance connections;then,the network parameters are updated after pruning is completed.The experimental results show that the proposed structure design method can effectively compress the scale of a DSCN model and improve its modeling speed;the model accuracy loss is small,and fine-tuning for accuracy restoration is not needed.The obtained DSCN model has certain application value in the field of regression analysis.
基金National Natural Science Foundation of China(Nos.61673262 and 50779033)National GF Basic Research Program(No.JCKY2021110B134)Fundamental Research Funds for the Central Universities。
文摘The multi-agent path planning problem presents significant challenges in dynamic environments,primarily due to the ever-changing positions of obstacles and the complex interactions between agents’actions.These factors contribute to a tendency for the solution to converge slowly,and in some cases,diverge altogether.In addressing this issue,this paper introduces a novel approach utilizing a double dueling deep Q-network(D3QN),tailored for dynamic multi-agent environments.A novel reward function based on multi-agent positional constraints is designed,and a training strategy based on incremental learning is performed to achieve collaborative path planning of multiple agents.Moreover,the greedy and Boltzmann probability selection policy is introduced for action selection and avoiding convergence to local extremum.To match radar and image sensors,a convolutional neural network-long short-term memory(CNN-LSTM)architecture is constructed to extract the feature of multi-source measurement as the input of the D3QN.The algorithm’s efficacy and reliability are validated in a simulated environment,utilizing robot operating system and Gazebo.The simulation results show that the proposed algorithm provides a real-time solution for path planning tasks in dynamic scenarios.In terms of the average success rate and accuracy,the proposed method is superior to other deep learning algorithms,and the convergence speed is also improved.
文摘AIM: To assess “top-down” treatment for deep remission of early moderate to severe Crohn’s disease (CD) by double balloon enteroscopy.
基金supported by the Universiti Tunku Abdul Rahman (UTAR) Malaysia under UTARRF (IPSR/RMC/UTARRF/2021-C1/T05)
文摘The recent surge of mobile subscribers and user data traffic has accelerated the telecommunication sector towards the adoption of the fifth-generation (5G) mobile networks. Cloud radio access network (CRAN) is a prominent framework in the 5G mobile network to meet the above requirements by deploying low-cost and intelligent multiple distributed antennas known as remote radio heads (RRHs). However, achieving the optimal resource allocation (RA) in CRAN using the traditional approach is still challenging due to the complex structure. In this paper, we introduce the convolutional neural network-based deep Q-network (CNN-DQN) to balance the energy consumption and guarantee the user quality of service (QoS) demand in downlink CRAN. We first formulate the Markov decision process (MDP) for energy efficiency (EE) and build up a 3-layer CNN to capture the environment feature as an input state space. We then use DQN to turn on/off the RRHs dynamically based on the user QoS demand and energy consumption in the CRAN. Finally, we solve the RA problem based on the user constraint and transmit power to guarantee the user QoS demand and maximize the EE with a minimum number of active RRHs. In the end, we conduct the simulation to compare our proposed scheme with nature DQN and the traditional approach.
基金Supported by the National Key Research and Development Plan(2019YFB1706401)。
文摘To optimize machine allocation and task dispatching in smart manufacturing factories, this paper proposes a manufacturing resource scheduling framework based on reinforcement learning(RL). The framework formulates the entire scheduling process as a multi-stage sequential decision problem, and further obtains the scheduling order by the combination of deep convolutional neural network(CNN) and improved deep Q-network(DQN). Specifically, with respect to the representation of the Markov decision process(MDP), the feature matrix is considered as the state space and a set of heuristic dispatching rules are denoted as the action space. In addition, the deep CNN is employed to approximate the state-action values, and the double dueling deep Qnetwork with prioritized experience replay and noisy network(D3QPN2) is adopted to determine the appropriate action according to the current state. In the experiments, compared with the traditional heuristic method, the proposed method is able to learn high-quality scheduling policy and achieve shorter makespan on the standard public datasets.
