Research on Data Anomaly Detection Method Based on Software-Defined Network for Power Grid

Smart Grid (SG) has promoted a new round of technological changes in the electric power field. As various smart devices are connected to the network, real-time collection of power data makes it possible to accurately grasp the operating status of the power grid. When managing huge data in smart grids, we face problems such as the inability of third-party cloud storage to meet data privacy requirements, the high cost of data sharing between different systems, and the lack of effective sharing incentive mechanisms. In view of the problem that third-party cloud storage solutions cannot meet the problem of data security storage and privacy protection, we studied a smart grid power data storage solution based on blockchain. Use hash calculation on the original power data to obtain a fixed-length hash summary, and then upload the hash summary to the blockchain for storage. The original power data and summary information are stored on IPFS off-chain. We designed smart contracts to implement user identity authentication and query log records to reflect the power data circulation process. In particular, we design a Byzantine Fault Tolerance protocol (Aggregate-signature Byzantine Fault Tolerance, ABFT) based on aggregate signatures. This protocol reduces communication complexity by adjusting the node broadcast method, and is more suitable for use in scenarios with many nodes. A large number of experimental results show that the solution we proposed can resist common attack types, and has a greater improvement in data storage overhead, average delay and throughput compared with similar solutions.