Research on Power Big Data Fusion Method Based on Graph Convolutional Neural Network

In recent years, in order to accelerate the construction of new power systems, power companies have carried out a series of information system construction, generating a massive amount of power related business data, which brings problems such as low query efficiency and inaccurate results to the real-time query performance of the system. Therefore, this article proposes a power big data fusion method based on graph convolutional neural networks, using association rule algorithms to horizontally connect various fragmented basic wide tables, indicators, and labels related to the perspective of power business, and aggregate them to form a logical business view. The association integrates more data information of power business objects, providing data support for power big data analysis, mining, and inference.