Privacy Protection And Anomaly Diagnosis Mechanism Of Secure Smart Grid Based On Distributed Ledger

As a new generation of new power system,smart grid achieves efficient transmission and distribution of power supply and demand,and is highly trusted by power users and power grid operators for its efficient management and quality service.However,frequent collection of users’ electricity consumption data through smart meters,data transmission through the open Internet,and mining and analysis of users’ fine-grained data all pose great threats to users’ privacy protection and data security.Therefore,it is urgent to study and design an efficient and reliable secure smart grid system.To address these problems,this paper mainly focuses on the following two aspects,one is the research on user communication data and privacy protection technology,the other is the research on reliable monitoring analysis and comprehensive processing technology of user power consumption data.In recent years,the research on information security and privacy protection of smart grid based on blockchain distributed ledger has attracted extensive attention from academia and industry.Currently,distributed ledger research can be divided into traditional forms(represented by blockchain)and interdisciplinary new forms(represented by distributed ledger based on directed acyclic graph(DAG)).Although traditional blockchain technology,which supports full order of transactions and smart contracts,is widely adopted,its scalability is limited by the chain organization and one-way extension of data,which to some extent makes it difficult to cope with excessive changes in transaction arrival rate of network overload,and even causes serious problems such as inability to reach consensus.In contrast,DAG distributed ledgers with high concurrency have great advantages in balancing decentralization and efficiency.However,the research on the consensus mechanism and ledger structure of DAG distributed ledger is still not mature.Moreover,neither of the two distributed ledger technologies can guarantee the authenticity of the user’s power consumption data.Artificial intelligence technology,represented by convolutional neural network(CNN),provides technical support for efficient intelligent computing of massive data,but there is little research on smart grid security diagnosis and intelligent analysis.In conclusion,firstly,we propose a lightweight and secure smart grid communication scheme based on the DAG distributed ledger,which achieves privacy preservation,data integrity,and high utility.Specifically,by integrating the homomorphic encryption,and the proxy-signature-like lightweight information authentication,the novel distributed solution for data aggregation is put forward to be able to protect the user’s data privacy,ensure communication confidentiality,and achieve source authentication of the communications.Secondly,through the performance optimization of smart grid,a trusted subjective logical credibility model is constructed,and combined with DAG distributed ledger technology,a new node consensus mechanism is studied and designed,which can efficiently store massive data of smart grid,and effectively solve the single point of failure and fair selection of consensus nodes.Meanwhile,by dynamically configuring the authority of the DAG distributed ledger and optimizing the Tips selection algorithm,the balance between the stability and security of the system is maximized.Thirdly,the smart grid system security diagnosis mechanism based on blockchain and convolutional neural network is studied and designed to realize the effective detection of malicious behavior and the "green" operation of smart grid.Specifically,by studying the overall framework of the combined convolutional neural network(CNN)security diagnosis system,the security diagnosis physical model is used to analyze the electricity consumption data and realize the effective detection of malicious behavior.Then,by constructing the specific function model of the security diagnosis system based on blockchain,a security diagnosis system based on block structure and smart contract is designed to ensure the safe,efficient and stable operation of smart power grid.Finally,extensive performance evaluations are conducted to illustrate that our study outperforms the state-of-the-art similar schemes in terms of computation complexity,communication cost,system scalability,transaction throughput,and response time(with low latency).