Dear Editor,The most serious heat related injury is exertional heat stroke(EHS).EHS occurs when healthy individuals perform physical activity in a hot and humid environment[1].A disrupted balance between heat producti...Dear Editor,The most serious heat related injury is exertional heat stroke(EHS).EHS occurs when healthy individuals perform physical activity in a hot and humid environment[1].A disrupted balance between heat production and dissipation in the human body results in excessive body heat storage in cases.It occurs frequently in the military population because of work characteristics such as the requirements to perform essential duties under prolonged heat stress,the need to achieve mission objectives during deployment operations,or the opportunities for training and selection for elite units[2].The pathophysiology of EHS is complex,which often results in thermoregulation failure,hemodynamic disturbance,and endotoxin release,and further causes multiple organ failure,probably increasing myocardial enzymes and N-terminal pro-brain natriuretic peptide(NT-proBNP)levels.Rhabdomyolysis caused by EHS often results from mechanical and metabolic injury to the striated muscle fibers accompanied with the release of muscle contents into the circulation[3].Liu et al.展开更多
Developing technologies that can be applied simultaneously in battery thermal management(BTM)and thermal runaway(TR)mitigation is significant to improving the safety of lithium-ion battery systems.Inorganic phase chan...Developing technologies that can be applied simultaneously in battery thermal management(BTM)and thermal runaway(TR)mitigation is significant to improving the safety of lithium-ion battery systems.Inorganic phase change material(PCM)with nonflammability has the potential to achieve this dual function.This study proposed an encapsulated inorganic phase change material(EPCM)with a heat transfer enhancement for battery systems,where Na_(2)HPO_(4)·12H_(2)O was used as the core PCM encapsulated by silica and the additive of carbon nanotube(CNT)was applied to enhance the thermal conductivity.The microstructure and thermal properties of the EPCM/CNT were analyzed by a series of characterization tests.Two different incorporating methods of CNT were compared and the proper CNT adding amount was also studied.After preparation,the battery thermal management performance and TR propagation mitigation effects of EPCM/CNT were further investigated on the battery modules.The experimental results of thermal management tests showed that EPCM/CNT not only slowed down the temperature rising of the module but also improved the temperature uniformity during normal operation.The peak battery temperature decreased from 76℃to 61.2℃at 2 C discharge rate and the temperature difference was controlled below 3℃.Moreover,the results of TR propagation tests demonstrated that nonflammable EPCM/CNT with good heat absorption could work as a TR barrier,which exhibited effective mitigation on TR and TR propagation.The trigger time of three cells was successfully delayed by 129,474 and 551 s,respectively and the propagation intervals were greatly extended as well.展开更多
In Beyond the Fifth Generation(B5G)heterogeneous edge networks,numerous users are multiplexed on a channel or served on the same frequency resource block,in which case the transmitter applies coding and the receiver u...In Beyond the Fifth Generation(B5G)heterogeneous edge networks,numerous users are multiplexed on a channel or served on the same frequency resource block,in which case the transmitter applies coding and the receiver uses interference cancellation.Unfortunately,uncoordinated radio resource allocation can reduce system throughput and lead to user inequity,for this reason,in this paper,channel allocation and power allocation problems are formulated to maximize the system sum rate and minimum user achievable rate.Since the construction model is non-convex and the response variables are high-dimensional,a distributed Deep Reinforcement Learning(DRL)framework called distributed Proximal Policy Optimization(PPO)is proposed to allocate or assign resources.Specifically,several simulated agents are trained in a heterogeneous environment to find robust behaviors that perform well in channel assignment and power allocation.Moreover,agents in the collection stage slow down,which hinders the learning of other agents.Therefore,a preemption strategy is further proposed in this paper to optimize the distributed PPO,form DP-PPO and successfully mitigate the straggler problem.The experimental results show that our mechanism named DP-PPO improves the performance over other DRL methods.展开更多
Data sharing and privacy protection are made possible by federated learning,which allows for continuous model parameter sharing between several clients and a central server.Multiple reliable and high-quality clients m...Data sharing and privacy protection are made possible by federated learning,which allows for continuous model parameter sharing between several clients and a central server.Multiple reliable and high-quality clients must participate in practical applications for the federated learning global model to be accurate,but because the clients are independent,the central server cannot fully control their behavior.The central server has no way of knowing the correctness of the model parameters provided by each client in this round,so clients may purposefully or unwittingly submit anomalous data,leading to abnormal behavior,such as becoming malicious attackers or defective clients.To reduce their negative consequences,it is crucial to quickly detect these abnormalities and incentivize them.In this paper,we propose a Federated Learning framework for Detecting and Incentivizing Abnormal Clients(FL-DIAC)to accomplish efficient and security federated learning.We build a detector that introduces an auto-encoder for anomaly detection and use it to perform anomaly identification and prevent the involvement of abnormal clients,in particular for the anomaly client detection problem.Among them,before the model parameters are input to the detector,we propose a Fourier transform-based anomaly data detectionmethod for dimensionality reduction in order to reduce the computational complexity.Additionally,we create a credit scorebased incentive structure to encourage clients to participate in training in order tomake clients actively participate.Three training models(CNN,MLP,and ResNet-18)and three datasets(MNIST,Fashion MNIST,and CIFAR-10)have been used in experiments.According to theoretical analysis and experimental findings,the FL-DIAC is superior to other federated learning schemes of the same type in terms of effectiveness.展开更多
针对Faster R-CNN算法对多目标、小目标检测精度不高的问题,本文提出一种基于Faster R-CNN的多任务增强裂缝图像检测(Multitask Enhanced Dam Crack Image Detection Based on Faster R-CNN,ME-Faster RCNN)方法。同时提出一种基于K-me...针对Faster R-CNN算法对多目标、小目标检测精度不高的问题,本文提出一种基于Faster R-CNN的多任务增强裂缝图像检测(Multitask Enhanced Dam Crack Image Detection Based on Faster R-CNN,ME-Faster RCNN)方法。同时提出一种基于K-means的多源自适应平衡TrAdaBoost的迁移学习方法(multi-source adaptive balance TrAdaBoost based on K-means,K-MABtrA)辅助网络训练,解决样本不足问题。ME-Faster R-CNN将图片输入ResNet-50网络提取特征;然后将所得特征图输入多任务增强RPN模型,同时改善RPN模型的锚盒尺寸和大小以提高检测识别精度,生成候选区域;最后将特征图和候选区域发送到检测处理网络。K-MABtrA方法利用K-means聚类删除与目标源差别较大的图像,再在多元自适应平衡TrAdaBoost迁移学习方法下训练模型。实验结果表明:将ME-Faster R-CNN在K-MABtrA迁移学习的条件下应用于小数据集大坝裂缝图像集的平均IoU为82.52%,平均精度mAP值为80.02%,与相同参数设置下的Faster R-CNN检测算法相比,平均IoU和mAP值分别提高了1.06%和1.56%。展开更多
基金supported by the Natural Science Foundation of Hainan Province(821QN389,821MS112,822MS198,820MS126,820QN383)the Military Medical Science and Technology Youth Incubation Program(20QNPY110,19QNP060)+6 种基金the Excellent Youth Incubation Program of Chinese PLA General Hospital(2020-YQPY-007)the Heatstroke Treatment and Research Center of Chinese PLA(413EGZ1D10)the Simulation Training for Treatment of Heatstroke,the Major Science and Technology Programme of Hainan Province(ZDKJ2019012)the National Key Research and Development Program of China(2018YFC2000400)the National S&T Resource Sharing Service Platform Project of China(YCZYPT[2018]07)the Specific Research Fund of Innovation Platform for Academicians of Hainan Province(YSPTZX202216)the Medical Big Data Research and Development Project of Chinese PLA General Hospital(MBD2018030).
文摘Dear Editor,The most serious heat related injury is exertional heat stroke(EHS).EHS occurs when healthy individuals perform physical activity in a hot and humid environment[1].A disrupted balance between heat production and dissipation in the human body results in excessive body heat storage in cases.It occurs frequently in the military population because of work characteristics such as the requirements to perform essential duties under prolonged heat stress,the need to achieve mission objectives during deployment operations,or the opportunities for training and selection for elite units[2].The pathophysiology of EHS is complex,which often results in thermoregulation failure,hemodynamic disturbance,and endotoxin release,and further causes multiple organ failure,probably increasing myocardial enzymes and N-terminal pro-brain natriuretic peptide(NT-proBNP)levels.Rhabdomyolysis caused by EHS often results from mechanical and metabolic injury to the striated muscle fibers accompanied with the release of muscle contents into the circulation[3].Liu et al.
基金financially supported by the National Key Research and Development Program(Grant No.2022YFE0207400)the National Natural Science Foundation of China(Grant No.U22A20168 and 52174225)。
文摘Developing technologies that can be applied simultaneously in battery thermal management(BTM)and thermal runaway(TR)mitigation is significant to improving the safety of lithium-ion battery systems.Inorganic phase change material(PCM)with nonflammability has the potential to achieve this dual function.This study proposed an encapsulated inorganic phase change material(EPCM)with a heat transfer enhancement for battery systems,where Na_(2)HPO_(4)·12H_(2)O was used as the core PCM encapsulated by silica and the additive of carbon nanotube(CNT)was applied to enhance the thermal conductivity.The microstructure and thermal properties of the EPCM/CNT were analyzed by a series of characterization tests.Two different incorporating methods of CNT were compared and the proper CNT adding amount was also studied.After preparation,the battery thermal management performance and TR propagation mitigation effects of EPCM/CNT were further investigated on the battery modules.The experimental results of thermal management tests showed that EPCM/CNT not only slowed down the temperature rising of the module but also improved the temperature uniformity during normal operation.The peak battery temperature decreased from 76℃to 61.2℃at 2 C discharge rate and the temperature difference was controlled below 3℃.Moreover,the results of TR propagation tests demonstrated that nonflammable EPCM/CNT with good heat absorption could work as a TR barrier,which exhibited effective mitigation on TR and TR propagation.The trigger time of three cells was successfully delayed by 129,474 and 551 s,respectively and the propagation intervals were greatly extended as well.
基金supported by the Key Research and Development Program of China(No.2022YFC3005401)Key Research and Development Program of China,Yunnan Province(No.202203AA080009,202202AF080003)Postgraduate Research&Practice Innovation Program of Jiangsu Province(No.KYCX21_0482).
文摘In Beyond the Fifth Generation(B5G)heterogeneous edge networks,numerous users are multiplexed on a channel or served on the same frequency resource block,in which case the transmitter applies coding and the receiver uses interference cancellation.Unfortunately,uncoordinated radio resource allocation can reduce system throughput and lead to user inequity,for this reason,in this paper,channel allocation and power allocation problems are formulated to maximize the system sum rate and minimum user achievable rate.Since the construction model is non-convex and the response variables are high-dimensional,a distributed Deep Reinforcement Learning(DRL)framework called distributed Proximal Policy Optimization(PPO)is proposed to allocate or assign resources.Specifically,several simulated agents are trained in a heterogeneous environment to find robust behaviors that perform well in channel assignment and power allocation.Moreover,agents in the collection stage slow down,which hinders the learning of other agents.Therefore,a preemption strategy is further proposed in this paper to optimize the distributed PPO,form DP-PPO and successfully mitigate the straggler problem.The experimental results show that our mechanism named DP-PPO improves the performance over other DRL methods.
基金supported by Key Research and Development Program of China (No.2022YFC3005401)Key Research and Development Program of Yunnan Province,China (Nos.202203AA080009,202202AF080003)+1 种基金Science and Technology Achievement Transformation Program of Jiangsu Province,China (BA2021002)Fundamental Research Funds for the Central Universities (Nos.B220203006,B210203024).
文摘Data sharing and privacy protection are made possible by federated learning,which allows for continuous model parameter sharing between several clients and a central server.Multiple reliable and high-quality clients must participate in practical applications for the federated learning global model to be accurate,but because the clients are independent,the central server cannot fully control their behavior.The central server has no way of knowing the correctness of the model parameters provided by each client in this round,so clients may purposefully or unwittingly submit anomalous data,leading to abnormal behavior,such as becoming malicious attackers or defective clients.To reduce their negative consequences,it is crucial to quickly detect these abnormalities and incentivize them.In this paper,we propose a Federated Learning framework for Detecting and Incentivizing Abnormal Clients(FL-DIAC)to accomplish efficient and security federated learning.We build a detector that introduces an auto-encoder for anomaly detection and use it to perform anomaly identification and prevent the involvement of abnormal clients,in particular for the anomaly client detection problem.Among them,before the model parameters are input to the detector,we propose a Fourier transform-based anomaly data detectionmethod for dimensionality reduction in order to reduce the computational complexity.Additionally,we create a credit scorebased incentive structure to encourage clients to participate in training in order tomake clients actively participate.Three training models(CNN,MLP,and ResNet-18)and three datasets(MNIST,Fashion MNIST,and CIFAR-10)have been used in experiments.According to theoretical analysis and experimental findings,the FL-DIAC is superior to other federated learning schemes of the same type in terms of effectiveness.
文摘针对Faster R-CNN算法对多目标、小目标检测精度不高的问题,本文提出一种基于Faster R-CNN的多任务增强裂缝图像检测(Multitask Enhanced Dam Crack Image Detection Based on Faster R-CNN,ME-Faster RCNN)方法。同时提出一种基于K-means的多源自适应平衡TrAdaBoost的迁移学习方法(multi-source adaptive balance TrAdaBoost based on K-means,K-MABtrA)辅助网络训练,解决样本不足问题。ME-Faster R-CNN将图片输入ResNet-50网络提取特征;然后将所得特征图输入多任务增强RPN模型,同时改善RPN模型的锚盒尺寸和大小以提高检测识别精度,生成候选区域;最后将特征图和候选区域发送到检测处理网络。K-MABtrA方法利用K-means聚类删除与目标源差别较大的图像,再在多元自适应平衡TrAdaBoost迁移学习方法下训练模型。实验结果表明:将ME-Faster R-CNN在K-MABtrA迁移学习的条件下应用于小数据集大坝裂缝图像集的平均IoU为82.52%,平均精度mAP值为80.02%,与相同参数设置下的Faster R-CNN检测算法相比,平均IoU和mAP值分别提高了1.06%和1.56%。
