With the continuous development of modern power system, more and more advanced metering devices have been deployed and popularized. A large amount of information is generated for power network analysis, load forecasti...
详细信息
ISBN:
(纸本)9798350303896
With the continuous development of modern power system, more and more advanced metering devices have been deployed and popularized. A large amount of information is generated for power network analysis, load forecasting, state perception and other aspects, providing information support for the related technical development. However, the increasing volume of data also brings huge pressure to the data storage, transmission, analysis and so on. Therefore, how to achieve datacompression while reducing the amount of data and retaining the characteristics of data information as much as possible has become a key concern. In this paper, a high-resolution power loaddatacompression method based on symbolic aggregate approximation (SAX) is adopted. The obtained power loaddata is first divided into event segment and non-event segment. The non-event segment is processed by SAX, and the event segments are divided into long event segments and short event segments. The original data of the short event segment is retained, and the long event segment is also processed by SAX. The comparison testing demonstrates that our proposed method achieves better performance. The average reconstruction accuracy obtained by RMSE index is 99.78%, and the average compression ratio is 21.90, which is promising for future engineering applications.
暂无评论