Time-Frequency Feature Combination Based Household Characteristic Identification Approach Using Smart Meter Data

Household characteristics play an important role in helping utilities carry out efficient and personalized services. Current methods to obtain such information, e.g., surveys, are usually costly and time-consuming. The widespread installation of smart meters enables the collection of fine-grained re...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on industry applications 2020-05, Vol.56 (3), p.2251-2262
Main Authors: Yan, Siqing, Li, Kangping, Wang, Fei, Ge, Xinxin, Lu, Xiaoxing, Mi, Zengqiang, Chen, Hongyu, Chang, Shengqiang
Format: Article
Language:English
Subjects:
Algorithms
Discrete Wavelet Transform
Discrete wavelet transforms
Electricity consumption
Electricity meters
Feature extraction
Frequency domain analysis
Garbage collection
Household characteristics
Households
Radio frequency
random forest (RF)
residential electricity consumption
Residential energy
smart meter data
Smart meters
support vector machine (SVM)
Support vector machines
Time-domain analysis
time–frequency features
Utilities
Wavelet transforms
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:Household characteristics play an important role in helping utilities carry out efficient and personalized services. Current methods to obtain such information, e.g., surveys, are usually costly and time-consuming. The widespread installation of smart meters enables the collection of fine-grained residential electricity consumption data and thus, making the identification of household characteristics from smart meter data possible. This article proposes a time-frequency feature combination based household characteristic identification approach using smart meter data. First, in addition to conventional time-domain statistical features, several frequency-domain features are extracted using discrete wavelet transform. Second, the random forest algorithm is used to select a subset of important features and remove redundant information contained in the original feature set. Third, a support vector machine is used as a classifier with the input of the selected features to infer the household characteristics. Finally, case study using the realistic data from Ireland indicates that the proposed approach shows better performance after incorporating the frequency-domain features.
ISSN:0093-9994
1939-9367
DOI:10.1109/TIA.2020.2981916