The modern power system has evolved into a cyber-physical system with deep coupling of physical and information domains,which brings new security risks.Aiming at the problem that the“information-physical”cross-domai...The modern power system has evolved into a cyber-physical system with deep coupling of physical and information domains,which brings new security risks.Aiming at the problem that the“information-physical”cross-domain attacks with key nodes as springboards seriously threaten the safe and stable operation of power grids,a risk propagation model considering key nodes of power communication coupling networks is proposed to study the risk propagation characteristics of malicious attacks on key nodes and the impact on the system.First,combined with the complex network theory,a topological model of the power communication coupling network is established,and the key nodes of the coupling network are screened out by Technique for Order Preference by Similarity to Ideal Solution(TOPSIS)method under the comprehensive evaluation index based on topological characteristics and physical characteristics.Second,a risk propagation model is established for malicious attacks on key nodes to study its propagation characteristics and analyze the state changes of each node in the coupled network.Then,two loss-causing factors:the minimum load loss ratio and transmission delay factor are constructed to quantify the impact of risk propagation on the coupled network.Finally,simulation analysis based on the IEEE 39-node system shows that the probability of node being breached(α)and the security tolerance of the system(β)are the key factors affecting the risk propagation characteristics of the coupled network,as well as the criticality of the node is positively correlated with the damage-causing factor.The proposed methodological model can provide an effective exploration of the diffusion of security risks in control systems on a macro level.展开更多
An advanced metering infrastructure(AMI)system plays a key role in the smart grid(SG),but it is vulnerable to cyberattacks.Current detection methods for AMI cyberattacks mainly focus on the data center or a distribute...An advanced metering infrastructure(AMI)system plays a key role in the smart grid(SG),but it is vulnerable to cyberattacks.Current detection methods for AMI cyberattacks mainly focus on the data center or a distributed independent node.On one hand,it is difficult to train an excellent detection intrusion model on a self-learning independent node.On the other hand,large amounts of data are shared over the network and uploaded to a central node for training.These processes may compromise data privacy,cause communication delay,and incur high communication costs.With these limitations,we propose an intrusion detection method for AMI system based on federated learning(FL).The intrusion detection system is deployed in the data concentrators for training,and only its model parameters are communicated to the data center.Furthermore,the data center distributes the learning to each data concentrator through aggregation and weight assignments for collaborative learning.An optimized deep neural network(DNN)is exploited for this proposed method,and extensive experiments based on the NSL-KDD dataset are carried out.From the results,this proposed method improves detection performance and reduces computation costs,communication delays,and communication overheads while guaranteeing data privacy.展开更多
基金This work was jointly supported by the National Natural Science Foundation of China(No.52177068)Hunan Provincial Natural Science Foundation of China(No.2023J30028)Graduate Research Innovation Project of Changsha University of Science and Technology(No.CXCLY2022076).
文摘The modern power system has evolved into a cyber-physical system with deep coupling of physical and information domains,which brings new security risks.Aiming at the problem that the“information-physical”cross-domain attacks with key nodes as springboards seriously threaten the safe and stable operation of power grids,a risk propagation model considering key nodes of power communication coupling networks is proposed to study the risk propagation characteristics of malicious attacks on key nodes and the impact on the system.First,combined with the complex network theory,a topological model of the power communication coupling network is established,and the key nodes of the coupling network are screened out by Technique for Order Preference by Similarity to Ideal Solution(TOPSIS)method under the comprehensive evaluation index based on topological characteristics and physical characteristics.Second,a risk propagation model is established for malicious attacks on key nodes to study its propagation characteristics and analyze the state changes of each node in the coupled network.Then,two loss-causing factors:the minimum load loss ratio and transmission delay factor are constructed to quantify the impact of risk propagation on the coupled network.Finally,simulation analysis based on the IEEE 39-node system shows that the probability of node being breached(α)and the security tolerance of the system(β)are the key factors affecting the risk propagation characteristics of the coupled network,as well as the criticality of the node is positively correlated with the damage-causing factor.The proposed methodological model can provide an effective exploration of the diffusion of security risks in control systems on a macro level.
基金supported in part by the National Natural Science Foundation of China(No.51807013)the Foundation of Hunan Educational Committee(No.18B137)+1 种基金the Research Project in Hunan Province Education Department(No.21C0577)Postgraduate Research and Innovation Project of Hunan Province,China(No.CX20210791)。
文摘An advanced metering infrastructure(AMI)system plays a key role in the smart grid(SG),but it is vulnerable to cyberattacks.Current detection methods for AMI cyberattacks mainly focus on the data center or a distributed independent node.On one hand,it is difficult to train an excellent detection intrusion model on a self-learning independent node.On the other hand,large amounts of data are shared over the network and uploaded to a central node for training.These processes may compromise data privacy,cause communication delay,and incur high communication costs.With these limitations,we propose an intrusion detection method for AMI system based on federated learning(FL).The intrusion detection system is deployed in the data concentrators for training,and only its model parameters are communicated to the data center.Furthermore,the data center distributes the learning to each data concentrator through aggregation and weight assignments for collaborative learning.An optimized deep neural network(DNN)is exploited for this proposed method,and extensive experiments based on the NSL-KDD dataset are carried out.From the results,this proposed method improves detection performance and reduces computation costs,communication delays,and communication overheads while guaranteeing data privacy.
