Power grid peak load clustering extraction method based on Pearson coefficient and MapReduce parallel computing

Abstract

A power grid peak load clustering extraction method based on Pearson coefficient and MapReduce parallel computing comprises the following steps: 1, conducting data cleaning and abnormal data restoration on a load data set, and then conducting dimensionality reduction on the load data set; 2, storing the data set subjected to dimension reduction in a Hadoop distributed file system line by line, anddividing the data set into slices to form sub-data sets; 3, reading each slice sub-data set in the step 2 by using a MapReduce computing architecture, and selecting an initial clustering center by using a Pearson correlation coefficient as a similarity criterion through a parallel model; and 4, allocating the clustering calculation task to the Map task node in the MapReduce, and combining the initial clustering center in the step 3 to complete parallel clustering of the data set. The method is more suitable for power load big data processing derived from more and more vigorous development information times.

Description

technical field [0001] The invention relates to a large data clustering algorithm based on a Map Reduce parallel framework, and is especially suitable for large-scale power grid load data sets during peak hours with complex calculations. Background technique [0002] With the vigorous development of smart grids and energy storage, massive power data are continuously generated in the operation, maintenance and management of the power grid, among which demand-side big data accounts for a large proportion. The planning and operation of smart grids require a good data foundation, so big data processing, load extraction and forecasting based on demand-side response need to be studied urgently. The demand-side big data includes high-dimensional and massive user daily / monthly load curves. Accurate analysis and research on these power consumption information data and obtaining corresponding load patterns can provide an important basis for decision-making of power grid companies. ...

AI Technical Summary

Problems solved by technology

[0007] The present invention overcomes the defects of traditional clustering techniques, and proposes a load curve clustering algorithm based on Pearson's correlation coefficient and combined with the Map Reduce parallel framework under the Hadoop platform. The problem of poor clustering effect and slow operation speed in the curve

Images