Lithium(Li) metal is considered as the most promising anode material for the next-generation high performance Li batteries.However,the uncontrollable dendritic growth impedes its commercial application.Herein,we desig...Lithium(Li) metal is considered as the most promising anode material for the next-generation high performance Li batteries.However,the uncontrollable dendritic growth impedes its commercial application.Herein,we design a 3 D Si@carbon nanofibers(CNFs)@ZnO-ZnO-Cu skeleton(SCZ) for guiding the homogeneous bottom-growth of Li metal.The top LixSi@CNFs and bottom LiyZn@CNFs layers could form conductivity and overpotential gradient to avoid the "top-growth" of Li metal.Moreover,the top lithiophilic LixSi@CNFs layer could regulate the nucleation and deposition of Li-ions even if the lithium dendrites grow out of the skeleton under high capacity Li deposition(30 mAh cm^(-2)).As a result,the SCZ-Li||LiFePO_(4) full cell delivers a high capacity of ~104 mAh g^(-1)(~94.82% capacity retention) after 2000 cycles at 5 C, elucidating the potential application of the 3 D double-gradient Li metal composite anode.展开更多
With the increasing awareness of privacy protection and the improvement of relevant laws,federal learning has gradually become a new choice for cross-agency and cross-device machine learning.In order to solve the prob...With the increasing awareness of privacy protection and the improvement of relevant laws,federal learning has gradually become a new choice for cross-agency and cross-device machine learning.In order to solve the problems of privacy leakage,high computational overhead and high traffic in some federated learning schemes,this paper proposes amultiplicative double privacymask algorithm which is convenient for homomorphic addition aggregation.The combination of homomorphic encryption and secret sharing ensures that the server cannot compromise user privacy from the private gradient uploaded by the participants.At the same time,the proposed TQRR(Top-Q-Random-R)gradient selection algorithm is used to filter the gradient of encryption and upload efficiently,which reduces the computing overhead of 51.78%and the traffic of 64.87%on the premise of ensuring the accuracy of themodel,whichmakes the framework of privacy protection federated learning lighter to adapt to more miniaturized federated learning terminals.展开更多
精确的电力负荷预测不仅在发电侧优化能源产能,而且在供给侧实现经济调度、绿色用电。调度工作能够根据预测负荷峰值科学有效的实现电网安全可靠运行。首先,提出了一种包含两个储备池的新型回声状态网预测方法,来预测每日电力负荷;其次...精确的电力负荷预测不仅在发电侧优化能源产能,而且在供给侧实现经济调度、绿色用电。调度工作能够根据预测负荷峰值科学有效的实现电网安全可靠运行。首先,提出了一种包含两个储备池的新型回声状态网预测方法,来预测每日电力负荷;其次,在传统的回声状态网的基础上,选取两个储备池进行串行连接,得到一种新型的回声状态网(Echo State Network with Double Reservoir,called DR-ESN)。DR-ESN能够更加有效的提取预测对象的特征信息,从而可以得到精度更高的预测结果;并利用批量梯度和岭回归算法来优化训练过程中的DR-ESN的6个参数。对广州市的实际用电量进行仿真,所得结果表明了预测方法的有效性。展开更多
The micro-electromechanical system(MEMS)infrared thermopile is the core working device of modern information detection systems such as spectrometers,gas sensors,and remote temperature sensors.We presented two differen...The micro-electromechanical system(MEMS)infrared thermopile is the core working device of modern information detection systems such as spectrometers,gas sensors,and remote temperature sensors.We presented two different structures of MEMS infrared thermopiles based on suspended film structures.They both deposited silicon nitride over the entire surface as a passivated absorber layer in place of a separate absorber zone,and the thermocouple strip was oriented in the same direction as the temperature gradient.The same MEMS preparation process was used and finally two different structures of the thermopile were characterized separately for testing to verify the impact of our design on the detector.The test results show that the circular and double-ended symmetrical thermopile detectors have responsivities of 27.932 V/W and 23.205 V/W,specific detectivities of 12.1×10^(7) cm·Hz^(1/2)·W^(-1) and 10.1×10^(7) cm·Hz^(1/2)·W^(-1),and response time of 26.2 ms and 27.06 ms,respectively.In addition,rectangular double-ended symmetric thermopile has a larger field of view than a circular thermopile detector,but is not as mechanically stable as a circular thermopile.展开更多
Device-to-device(D2D)communications underlying cellular networks enabled by unmanned aerial vehicles(UAV)have been regarded as promising techniques for next-generation communications.To mitigate the strong interferenc...Device-to-device(D2D)communications underlying cellular networks enabled by unmanned aerial vehicles(UAV)have been regarded as promising techniques for next-generation communications.To mitigate the strong interference caused by the line-of-sight(LoS)airto-ground channels,we deploy a reconfigurable intelligent surface(RIS)to rebuild the wireless channels.A joint optimization problem of the transmit power of UAV,the transmit power of D2D users and the RIS phase configuration are investigated to maximize the achievable rate of D2D users while satisfying the quality of service(QoS)requirement of cellular users.Due to the high channel dynamics and the coupling among cellular users,the RIS,and the D2D users,it is challenging to find a proper solution.Thus,a RIS softmax deep double deterministic(RIS-SD3)policy gradient method is proposed,which can smooth the optimization space as well as reduce the number of local optimizations.Specifically,the SD3 algorithm maximizes the reward of the agent by training the agent to maximize the value function after the softmax operator is introduced.Simulation results show that the proposed RIS-SD3 algorithm can significantly improve the rate of the D2D users while controlling the interference to the cellular user.Moreover,the proposed RIS-SD3 algorithm has better robustness than the twin delayed deep deterministic(TD3)policy gradient algorithm in a dynamic environment.展开更多
基金financial support from the National Natural Science Foundation of China(Grant Nos.51701169,51871188 and 51931006)the National Key R&D Program of China(Grant No.2016YFA0202602)+1 种基金the Natural Science Foundation of Fujian Province of China(No.2019J06003)the "Double-First Class" Foundation of Materials and Intelligent Manufacturing Discipline of Xiamen University。
文摘Lithium(Li) metal is considered as the most promising anode material for the next-generation high performance Li batteries.However,the uncontrollable dendritic growth impedes its commercial application.Herein,we design a 3 D Si@carbon nanofibers(CNFs)@ZnO-ZnO-Cu skeleton(SCZ) for guiding the homogeneous bottom-growth of Li metal.The top LixSi@CNFs and bottom LiyZn@CNFs layers could form conductivity and overpotential gradient to avoid the "top-growth" of Li metal.Moreover,the top lithiophilic LixSi@CNFs layer could regulate the nucleation and deposition of Li-ions even if the lithium dendrites grow out of the skeleton under high capacity Li deposition(30 mAh cm^(-2)).As a result,the SCZ-Li||LiFePO_(4) full cell delivers a high capacity of ~104 mAh g^(-1)(~94.82% capacity retention) after 2000 cycles at 5 C, elucidating the potential application of the 3 D double-gradient Li metal composite anode.
基金supported by the National Natural Science Foundation of China(Grant Nos.62172436,62102452)the National Key Research and Development Program of China(2023YFB3106100,2021YFB3100100)the Natural Science Foundation of Shaanxi Province(2023-JC-YB-584).
文摘With the increasing awareness of privacy protection and the improvement of relevant laws,federal learning has gradually become a new choice for cross-agency and cross-device machine learning.In order to solve the problems of privacy leakage,high computational overhead and high traffic in some federated learning schemes,this paper proposes amultiplicative double privacymask algorithm which is convenient for homomorphic addition aggregation.The combination of homomorphic encryption and secret sharing ensures that the server cannot compromise user privacy from the private gradient uploaded by the participants.At the same time,the proposed TQRR(Top-Q-Random-R)gradient selection algorithm is used to filter the gradient of encryption and upload efficiently,which reduces the computing overhead of 51.78%and the traffic of 64.87%on the premise of ensuring the accuracy of themodel,whichmakes the framework of privacy protection federated learning lighter to adapt to more miniaturized federated learning terminals.
文摘精确的电力负荷预测不仅在发电侧优化能源产能,而且在供给侧实现经济调度、绿色用电。调度工作能够根据预测负荷峰值科学有效的实现电网安全可靠运行。首先,提出了一种包含两个储备池的新型回声状态网预测方法,来预测每日电力负荷;其次,在传统的回声状态网的基础上,选取两个储备池进行串行连接,得到一种新型的回声状态网(Echo State Network with Double Reservoir,called DR-ESN)。DR-ESN能够更加有效的提取预测对象的特征信息,从而可以得到精度更高的预测结果;并利用批量梯度和岭回归算法来优化训练过程中的DR-ESN的6个参数。对广州市的实际用电量进行仿真,所得结果表明了预测方法的有效性。
基金This work was supported in part by the National Natural Science Foundation of China(Grant No.51935011)Innovative Research Group Project of National Science Foundation of China(Grant No.51821003)+1 种基金Fund for Shanxi“1331 Project”Key Subject Construction,Key Research and Development Project of Shanxi Province(Grant Nos.202102030201001 and 202102030201009)Key Special Project of Science and Technology of Shanxi Province(Grant No.202201030201004).
文摘The micro-electromechanical system(MEMS)infrared thermopile is the core working device of modern information detection systems such as spectrometers,gas sensors,and remote temperature sensors.We presented two different structures of MEMS infrared thermopiles based on suspended film structures.They both deposited silicon nitride over the entire surface as a passivated absorber layer in place of a separate absorber zone,and the thermocouple strip was oriented in the same direction as the temperature gradient.The same MEMS preparation process was used and finally two different structures of the thermopile were characterized separately for testing to verify the impact of our design on the detector.The test results show that the circular and double-ended symmetrical thermopile detectors have responsivities of 27.932 V/W and 23.205 V/W,specific detectivities of 12.1×10^(7) cm·Hz^(1/2)·W^(-1) and 10.1×10^(7) cm·Hz^(1/2)·W^(-1),and response time of 26.2 ms and 27.06 ms,respectively.In addition,rectangular double-ended symmetric thermopile has a larger field of view than a circular thermopile detector,but is not as mechanically stable as a circular thermopile.
基金supported by the National Natural Science Foundation of China under Grant Nos.62201462 and 62271412.
文摘Device-to-device(D2D)communications underlying cellular networks enabled by unmanned aerial vehicles(UAV)have been regarded as promising techniques for next-generation communications.To mitigate the strong interference caused by the line-of-sight(LoS)airto-ground channels,we deploy a reconfigurable intelligent surface(RIS)to rebuild the wireless channels.A joint optimization problem of the transmit power of UAV,the transmit power of D2D users and the RIS phase configuration are investigated to maximize the achievable rate of D2D users while satisfying the quality of service(QoS)requirement of cellular users.Due to the high channel dynamics and the coupling among cellular users,the RIS,and the D2D users,it is challenging to find a proper solution.Thus,a RIS softmax deep double deterministic(RIS-SD3)policy gradient method is proposed,which can smooth the optimization space as well as reduce the number of local optimizations.Specifically,the SD3 algorithm maximizes the reward of the agent by training the agent to maximize the value function after the softmax operator is introduced.Simulation results show that the proposed RIS-SD3 algorithm can significantly improve the rate of the D2D users while controlling the interference to the cellular user.Moreover,the proposed RIS-SD3 algorithm has better robustness than the twin delayed deep deterministic(TD3)policy gradient algorithm in a dynamic environment.
