Geoword-based multivariate power time series data indexing method

Abstract

The invention relates to a Geoword-based multivariate power time series data indexing method. The Geoword-based multivariate power time series data indexing method comprises the following steps of obtaining original multivariate power time series data, and performing dimension reduction on the original multivariate power time series data; conducting Geoword encoding on dimension-reduced multivariate power time series data, and obtaining an MTSAX expression of the multivariate power time series data; constructing MTSAX data indexes of a multivariate power time series according to the MTSAX expression. Compared with the prior art, the Geoword-based multivariate power time series data indexing method can conduct dynamic division, specifies the precision, has no overlapped nodes and the like.

Description

technical field [0001] The invention relates to the field of power data processing, in particular to a Geoword-based multivariate power time-series data indexing method. Background technique [0002] User electricity load data is a kind of massive time series data, which has the characteristics of large user scale, high data collection density, and close correlation with a large amount of economic and social data. Time series data indexing technology is very important to reduce the time cost of data query and retrieval and improve the efficiency of time series mining (such as classification, clustering, abnormal point monitoring, pattern discovery, etc.). Time series is a series of data arranged in chronological order. According to the number of variables contained in the time series, it can be divided into univariate time series and multivariate time series. The user's electricity load data contains multiple original information such as daily electricity consumption, volta...

AI Technical Summary

Problems solved by technology

[0007] The current method based on space division retains the approximate information of the multivariate time series, but basically divides the space fixedly. For the index, the data distribution may not be uniform, which will reduce the query efficiency

Indexes based on feature compression find similar time series through dimensionality reduction, but usually cannot preserve the original multivariate time series information

Indexes based on metric spaces use lower-bound filtering techniques such as triangle inequality to improve query efficiency, but such indexes are still difficult to overcome the "dimension disaster problem" faced by high-dimensional data indexes

Images