Event-triggered Based Resilient Distributed Energy Management Of Micro-grid Under Cyber Attacks

With the flourishing development of society,the demand for electricity in society is increasing.Microgrid,as a new type of energy management method,refers to a small-scale power grid composed of various energy sources and energy storage devices,and has attracted more and more attention from people.The economic operation of microgrid is an important issue of concern to many researchers.Traditional economic dispatch problem(EDP)focuses on the generation side,and minimizes the total generation cost to satisfy the physical constraints of each generator.In contrast,the goal of demand response(DR)is to maximize the overall benefits of consumers on the demand side.In most existing research works,these two operations are studied separately.The energy management problem combines both,considering the interests of all participants in the microgrid,and maximizing the social welfare sum of all generation units and load units.In addition,in recent years,network attack incidents have occurred frequently,and malicious data attacks against hostile intelligent agents pose a significant threat to information-physical fusion systems,which also applies to energy management of microgrids.Therefore,designing a reliable,stable,efficient,and environmentally friendly energy management system provides a solid guarantee for the operation of microgrid.As an important link in the microgrid,it has significant practical significance.This paper discusses the design of a distributed optimization algorithm that is resilient to network attacks for the energy management problem in microgrids.The algorithm is based on an event-triggered mechanism and is designed to assign different resources to three types of nodes in the microgrid: trusted nodes,ordinary nodes,and malicious nodes.The trusted nodes are given more resources to ensure high security and prevent attacks or natural failures.The trusted nodes then form a connected dominating set in the network topology,which is used to filter out false states propagated by attackers and reduce the communication burden.The algorithm’s convergence is theoretically proven to converge to the optimal solution for any initial state of trusted and normal nodes.Finally,the algorithm is tested on 9 agents,and the results demonstrate that it is effective in reducing network attacks and communication burden.Next,we consider the distributed energy management of microgrids with transmission losses and network attacks.Considering the existence of malicious attacks in the network,we first establish a social welfare maximization model with transmission losses.Then,based on a three-node defense strategy and an event-triggered mechanism,we propose an elastic microgrid energy management algorithm based on an event-triggered mechanism and provide a rigorous convergence analysis.In order to verify the effectiveness and scalability of the algorithm proposed in this paper,case studies were conducted on an 8-node test system and the IEEE 39-bus system.The results show that the proposed algorithm performs well in both test systems and can converge the remaining nodes to the optimal electricity price and optimal output power while greatly reducing the network communication burden even when there are malicious attacks in the network.Finally,the energy management algorithm proposed in this paper under network attack is compared with the algorithm without attack.The results show that the proposed algorithm can still perform almost the same as the algorithm without attack,which also indicates the correctness of all algorithms.