1. Load morphological analysis method for demand-side user based on multi-stage clustering
- Author
-
Li Xilan, Shengzhou Ke, Dishan Yang, Xiang Kangli, and Tongyu Yan
- Subjects
Demand side ,business.industry ,Computer science ,Cosine similarity ,Process (computing) ,Pattern recognition ,Measure (mathematics) ,Multi stage ,ComputingMethodologies_PATTERNRECOGNITION ,Similarity (network science) ,Morphological analysis ,Artificial intelligence ,business ,Cluster analysis - Abstract
In this paper, a demand-side user load morphological analysis method based on multi-stage clustering is proposed to improve the efficiency of traditional user load morphological analysis method when the number of users increases dramatically. This method measures the difference of user load curve by cosine similarity, and divides user load morphological clustering into three layers, which are the process of agglomeration-clustering-recovery. Respectively, so as to effectively reduce the influence of increasing number of users on the clustering effect. The comparison with the traditional method shows that this method can effectively improve the efficiency and accuracy of load morphological analysis method, and can effectively measure the similarity between load curves, which has the use-value in demand side management.
- Published
- 2018