Parallelization clustering method for power communication data resources

Abstract

The invention belongs to the technical field of power systems and more specifically relates to a parallelization clustering method for power communication data resources. First, point density of different sample points in a data set is calculated in parallel; k initial clustering centers are selected according to the size of the point density and the mutual distance; finally, a k-medoids algorithmis utilized for non-center point distribution and center point update on a Hadoop platform in parallel until the center point and the non-center point of each cluster do not change any more and the final clustering result is obtained. According to the invention, the initial points of the clustering algorithm are selected through combination of the size of the point density calculated according tothe sample points and the mutual distance, the initial cluster center selection mechanism is improved, the clustering iteration operation effectiveness is improved, the search range is reduced and the clustering effect is improved; parallelization resource allocation is optimized, clustering time is shortened and clustering precision is improved, effective analysis and utilization of power data are well supported and the comprehensive benefits of power communication network related data are developed.

Description

technical field [0001] The invention belongs to the technical field of power systems, and more particularly, relates to a parallel clustering method for power communication data resources. Background technique [0002] With the continuous development of the power communication network with the function as the center, a large number of operation and maintenance management systems are produced which are divided into specialties, functions and management domains, which in turn leads to the generation of "island phenomenon" of power operation and maintenance data, which seriously inhibits a large number of scattered The comprehensive benefits of power communication operation and maintenance data are brought into play. How to use distributed systems to better handle these huge and complex power communication operation and maintenance data has become a hot research issue. As an effective means of data processing, cluster analysis supports the integration and classification of a l...

AI Technical Summary

Problems solved by technology

[0003] The quality of the initial point selection in the clustering algorithm will have a huge impact on the accuracy and efficiency of the entire clustering algorithm. In addition, the resource allocation schemes of different parallel algorithms will also affect the execution time of the clustering algorithm.

In the analysis of existing patent literature, many clustering algorithms do not optimize the selection of initial points, which has a great impact on the mechanism and clustering speed; in addition, simple random sampling is performed on the original data to complete the data segmentation. The execution efficiency of data initialization is not high; there is also a hierarchical clustering method based on graph generation, which implements a parallel clustering method under the MapReduce parallel framework. The accuracy is not ideal and has certain limitations

Images