Load profile disaggregation by Blind source separation: A wavelets-assisted independent component analysis approach

Author

Yongli Zhu ; Songtao Lu

Author_Institution

ECE Dept., Univ. of Tennessee, Knoxville, TN, USA

fYear

2014

fDate

27-31 July 2014

Firstpage

1

Lastpage

5

Abstract

In this paper, a Blind-source separation method, i.e. Independent Component Analysis (ICA) is used for disaggregating the substation load profile into different patterns, i.e. residential and commercial groups. The smart meter data from a down town substation has been used. Principle Component Analysis (PCA) is applied for data reduction. Wavelet analysis is used to extract the trend signal from the original load profile as inputs for the ICA routine. Final results verify the effectiveness of this load profile disaggregation approach.

Keywords

blind source separation; data communication; independent component analysis; load forecasting; principal component analysis; smart meters; substations; ICA; PCA; blind source separation method; commercial group pattern; independent component analysis; load forecasting; principle component analysis; residential group pattern; signal extraction; smart meter data reduction; substation load profile disaggregation; wavelet analysis; Cities and towns; Independent component analysis; Market research; Principal component analysis; Substations; Wavelet analysis; Wavelet transforms; Blind-source separation; load modeling; load profile; smart meter; statistic learning;

fLanguage

English

Publisher

ieee

Conference_Titel

PES General Meeting | Conference & Exposition, 2014 IEEE

Conference_Location

National Harbor, MD

Type

conf

DOI

10.1109/PESGM.2014.6938947

Filename

6938947