Time series data compression and recovery method based on power system

Abstract

The invention provides a time series data compression and recovery method based on a power system used for Data compression and recovery on numeric data. Based on a most basic compression method as anexample, the compression method comprises the steps that an edge computing device stores time series data collected by a power terminal; the edge computing device classifies the time series data according to the data characteristics and divides the time series data into stable data and fluctuation data; the stable data is compressed and stored by adopting an improved DEGA algorithm; and the fluctuation data is compressed and stored by using an automatic encoder in a neural network; The time series data is processed by utilizing equipartition codes, the data coding time is shortened, and the time series data are classified and processed by adopting an automatic encoder and the improved DEGA algorithm, so that the data compression ratio is higher.

Description

technical field [0001] The invention proposes a time series data compression and recovery method based on the power system, which mainly performs data compression and recovery for numerical data in the power system. Background technique [0002] With the rapid development of technology in the field of communication and Internet, people put forward new requirements for the intelligence and automation of power systems, which makes the types of data generated by various intelligent terminal equipment continue to increase. However, the disconnection of data between intelligent power devices creates an "island effect" among the data, which makes data collection and analysis face a huge challenge. The rise of edge computing has effectively solved this problem. At the edge of the network close to the object or data source, an open platform integrating connection, computing, storage and application is needed to provide edge intelligent services for the data of power sensor node devi...

AI Technical Summary

Problems solved by technology

Lossy compression has its advantages, but in grid data, some details of the data may be discarded during the compression process, resulting in the original data set cannot be fully recovered from the compressed version

The compression of lossless data solves the problems existing in lossy data. It is generally aimed at data compression of text, audio and video types. Related compression coding methods include Shannon-Fanno coding, Huffman coding and other algorithms. These technologies are through Calculate the probability of characters in the text and then compress the data, but the time efficiency of compression and recovery of numerical data (power data) is low

Images