Anomaly Detection Method for Multi-Source Data in Smart Grid Based on Big Data Technology
This paper proposes an automatic detection method for multi-source data in smart grids based on big data technology.The method involves unified modeling of multi-source heterogeneous data from the power grid,extracting key physical quantities and state information to construct an association graph spectrum of multi-source data.Subsequently,a graph convolutional network(GCN)model is designed based on the association graph spectrum,which learns end-to-end the complex associative features between multi-source data to achieve automatic detection of anomalies such as equipment failures.Simulation experiments demonstrate that the method fully exploits the intrinsic connections of multi-source data,significantly improving the accuracy of anomaly detection,and providing a basis for power grid monitoring and decision-making.
big data technologysmart gridmulti-source dataautomatic detection
万方数据
维普
