现代化电力系统向规模化和广域化发展,区部控制已不能完全实现整个系统的稳定运行,使得电力系统开始依赖开放型通信网络,从而系统中通信时滞的问题越发严重。通过考虑通信时延和可再生能源波动,该文建立新型的集成系统模型,然后提出负...现代化电力系统向规模化和广域化发展,区部控制已不能完全实现整个系统的稳定运行,使得电力系统开始依赖开放型通信网络,从而系统中通信时滞的问题越发严重。通过考虑通信时延和可再生能源波动,该文建立新型的集成系统模型,然后提出负载频率控制(load frequency control,LFC)以减小频率偏差。同时,采用滑模(sliding mode,SM)方法对LFC进行优化,可以提高不确定性和通信延迟不匹配的可再生系统的稳定性。此外,风力发电机组的输出功率还可以基于集成模型主动响应系统频率变化。最后,通过使用RTDS实验装置有效地证明该文提出的控制策略可以在各种负载扰动和通信延迟运行条件下来减小频率波动。展开更多
针对通过无线通信网络实现远程控制的地面无人系统,分析了地面无人系统在工作的过程中,网络时延对系统的影响。基于网络时延的分布特性,提出了一种贝叶斯算法(Bayesian algorithm,BO)优化的长短期记忆(long-term and short-term memory,...针对通过无线通信网络实现远程控制的地面无人系统,分析了地面无人系统在工作的过程中,网络时延对系统的影响。基于网络时延的分布特性,提出了一种贝叶斯算法(Bayesian algorithm,BO)优化的长短期记忆(long-term and short-term memory,LSTM)神经网络时延预测模型,在Matlab软件中搭建了该模型,并通过网络时延训练集数据对模型进行了训练,在网络时延测试集数据上对训练好的模型进行了测试,最后,就R2、RMSE和MAE评价指标对测试效果和GRU、BO-GRU以及LSTM预测方法进行了对比,结果表明,BO算法优化的LSTM神经网络时延预测模型RMSE和MAE评价结果更低,预测精度更高,预测效果更好,验证了提出的网络时延预测模型的有效性。展开更多
In this paper,an optimal secondary control strategy is proposed for islanded AC microgrids considering communi-cation time-delays.The proposed method is designed based on the data-driven principle,which consists of an...In this paper,an optimal secondary control strategy is proposed for islanded AC microgrids considering communi-cation time-delays.The proposed method is designed based on the data-driven principle,which consists of an offine training phase and online application phase.For offline training,each control agent is formulated by a deep neural network(DNN)and trained based on a multi-agent deep reinforcement learning(MA-DRL)framework.A deep deterministic policy gradient(DDPG)algorithm is improved and applied to search for an optimal policy of the secondary control,where a global cost function is developed to evaluate the overall control performance.In addition,the communication time-delay is introduced in the system to enrich training scenarios,which aims to solve the time-delay problem in the secondary control.For the online stage,each controller is deployed in a distributed way which only requires local and neighboring information for each DG.Based on this,the well-trained controllers can provide optimal solutions under load variations,and communication time-delays for online applications.Several case studies are conducted to validate the feasibility and stability of the proposed secondary control.Index Terms-Communication time-delay,global cost function,islanded AC microgrid,multi-agent deep reinforcement learning(MA-DRL),secondary control.展开更多
This paper analyzes the multi-cluster flocking behavior of a Cucker-Smale model involving delays and a short-range communication weight.In each sub-flocking group,the velocity between agents is alignment and the posit...This paper analyzes the multi-cluster flocking behavior of a Cucker-Smale model involving delays and a short-range communication weight.In each sub-flocking group,the velocity between agents is alignment and the position locates at a limited domain;but in different sub-flocking groups,the position between agents is unbounded.By constructing dissipative differential inequalities of subensembles together with Lyapunov functional methods,the authors provide the sufficient condition for the multi-cluster flocking emerging.The sufficient condition includes the estimation of the range of coupling strength and the upper bound of time delay.As a result,the authors show that the coupling strength among agents and initial threshold value determine the multi-cluster flocking behavior of the delayed Cucker-Smale model.展开更多
文摘现代化电力系统向规模化和广域化发展,区部控制已不能完全实现整个系统的稳定运行,使得电力系统开始依赖开放型通信网络,从而系统中通信时滞的问题越发严重。通过考虑通信时延和可再生能源波动,该文建立新型的集成系统模型,然后提出负载频率控制(load frequency control,LFC)以减小频率偏差。同时,采用滑模(sliding mode,SM)方法对LFC进行优化,可以提高不确定性和通信延迟不匹配的可再生系统的稳定性。此外,风力发电机组的输出功率还可以基于集成模型主动响应系统频率变化。最后,通过使用RTDS实验装置有效地证明该文提出的控制策略可以在各种负载扰动和通信延迟运行条件下来减小频率波动。
文摘针对通过无线通信网络实现远程控制的地面无人系统,分析了地面无人系统在工作的过程中,网络时延对系统的影响。基于网络时延的分布特性,提出了一种贝叶斯算法(Bayesian algorithm,BO)优化的长短期记忆(long-term and short-term memory,LSTM)神经网络时延预测模型,在Matlab软件中搭建了该模型,并通过网络时延训练集数据对模型进行了训练,在网络时延测试集数据上对训练好的模型进行了测试,最后,就R2、RMSE和MAE评价指标对测试效果和GRU、BO-GRU以及LSTM预测方法进行了对比,结果表明,BO算法优化的LSTM神经网络时延预测模型RMSE和MAE评价结果更低,预测精度更高,预测效果更好,验证了提出的网络时延预测模型的有效性。
基金supported by the Ministry of Education(MOE),Republic of Singapore,under grant(AcRFTIER-1 RT11/22)。
文摘In this paper,an optimal secondary control strategy is proposed for islanded AC microgrids considering communi-cation time-delays.The proposed method is designed based on the data-driven principle,which consists of an offine training phase and online application phase.For offline training,each control agent is formulated by a deep neural network(DNN)and trained based on a multi-agent deep reinforcement learning(MA-DRL)framework.A deep deterministic policy gradient(DDPG)algorithm is improved and applied to search for an optimal policy of the secondary control,where a global cost function is developed to evaluate the overall control performance.In addition,the communication time-delay is introduced in the system to enrich training scenarios,which aims to solve the time-delay problem in the secondary control.For the online stage,each controller is deployed in a distributed way which only requires local and neighboring information for each DG.Based on this,the well-trained controllers can provide optimal solutions under load variations,and communication time-delays for online applications.Several case studies are conducted to validate the feasibility and stability of the proposed secondary control.Index Terms-Communication time-delay,global cost function,islanded AC microgrid,multi-agent deep reinforcement learning(MA-DRL),secondary control.
基金supported by the National Natural Science Foundation of China under Grant Nos.11671011and 11428101。
文摘This paper analyzes the multi-cluster flocking behavior of a Cucker-Smale model involving delays and a short-range communication weight.In each sub-flocking group,the velocity between agents is alignment and the position locates at a limited domain;but in different sub-flocking groups,the position between agents is unbounded.By constructing dissipative differential inequalities of subensembles together with Lyapunov functional methods,the authors provide the sufficient condition for the multi-cluster flocking emerging.The sufficient condition includes the estimation of the range of coupling strength and the upper bound of time delay.As a result,the authors show that the coupling strength among agents and initial threshold value determine the multi-cluster flocking behavior of the delayed Cucker-Smale model.
