High-Performance Fusion Storage Technology for Massive Power Grid Dispatching and Operation Data

With the progression of new type power system infrastructures, grid control and operational data have begun to exhibit characteristics of voluminousness, multi-source origins, heterogeneity, and stringent real-time demands. Conventional data storage and processing methodologies are increasingly inadequate to address these requirements. This paper introduces a high-performance storage and processing framework for the extensive operational data of grid control, leveraging the integration of data lake and data warehouse technologies. By implementing stratified storage and refined scheduling algorithms, the framework achieves enhanced real-time throughput and superior processing capabilities for massive datasets. The proposed solution employs data lakes for the retention of raw data and data warehouses for the storage of refined and transformed data, integrating both streaming and batch computing to fulfill the real-time and precision exigencies of grid control. Empirical results demonstrate that the proposed framework significantly augments data storage and processing efficiency, thereby furnishing robust support for grid control operations.