Loading…

An Adaptive Weighted Pearson Similarity Measurement Method for Load Curve Clustering

Load curve data from advanced metering infrastructure record the consumers’ behavior. User consumption models help one understand a more intelligent power provisioning and clustering the load data is one of the popular approaches for building these models. Similarity measurements are important in th...

Full description

Saved in:
Bibliographic Details
Published in:Energies (Basel) 2018-09, Vol.11 (9), p.2466
Main Authors: Lin, Rongheng, Wu, Budan, Su, Yun
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Load curve data from advanced metering infrastructure record the consumers’ behavior. User consumption models help one understand a more intelligent power provisioning and clustering the load data is one of the popular approaches for building these models. Similarity measurements are important in the clustering model, but, load curve data is a time series style data, and traditional measurement methods are not suitable for load curve data. To cluster the load curve data more accurately, this paper applied an enhanced Pearson similarity for load curve data clustering. Our method introduces the ‘trend alteration point’ concept and integrates it with the Pearson similarity. By introducing a weight for Pearson distance, this method helps to keep the whole contour of the load data and the partial similarity. Based on the weighed Pearson distance, a weighed Pearson-based hierarchy clustering algorithm is proposed. Years of load curve data are used for evaluation. Several user consumption models are found and analyzed. Results show that the proposed method improves the accuracy of load data clustering.
ISSN:1996-1073
1996-1073
DOI:10.3390/en11092466