A Load Curve Integrated Spectral Clustering Method Considering Dual-Scale Similarity

Abstract

The invention discloses a load curve integrated spectrum clustering method considering the dual-scale similarity. The method calculates the similarity of the load shape change through the cosine distance of the load difference vector, and constructs a dual-scale similarity measurement method for measuring the similarity of the load; through Integrated clustering achieves clustering performance improvement and the effective combination of two measurement methods. Spectral clustering is used as the base clustering model generation method. By selecting different similarity measurement methods, setting different cluster numbers and running randomly to ensure the base clustering Class diversity, with weighted consistency matrix and spectral clustering as the clustering integration strategy, the Davidson-Pauding index DBI or the new index MDBI is used as the clustering evaluation index in the clustering integration process, and the reciprocal of DBI or MDBI is used as the weight The adaptation setting is based on the calculation of the consistency matrix, and then the final ensemble clustering is achieved by spectral clustering. This method has excellent clustering effectiveness and robustness, and can avoid the defect that a single spectral clustering method requires parameter optimization for different data sets.

Description

technical field [0001] The invention relates to a load curve integrated spectrum clustering method considering double-scale similarity, and belongs to the field of load characteristic analysis of electric power systems. Background technique [0002] Under the background of urban energy Internet, the construction and improvement of electricity collection information system and dispatching, transportation inspection, and marketing business systems have promoted the rapid accumulation of electric power data resources. Valuable information such as energy consumption characteristics is hidden in power data, which needs to be mined by applying data analysis technology. As an unsupervised learning technology, clustering is suitable for the classification of unlabeled load curves, providing power companies with classification results based on differences in load characteristics, helping power companies to accurately grasp the laws of user energy consumption behaviors, and providing ...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem of summarization in the background technology, the present invention provides a load curve integrated spectrum clustering method considering dual-scale similarity, which combines the advantages of dual-scale similarity and integrated clustering, and constructs a load curve considering the distance and An ensemble spectral clustering model for differences in morphological properties

Images