Loading…
Anomaly Detection Using Power Signature of Consumer Electrical Devices
The use of the smart grid for developing intelligent applications is a current trend of great importance. One advantage lies in the possibility of direct monitoring of all devices connected to the electrical network in order to prevent possible malfunctions. Therefore, this paper proposes a method f...
Saved in:
Published in: | Advances in electrical and computer engineering 2015-01, Vol.15 (1), p.89-94 |
---|---|
Main Authors: | , |
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!
|
Summary: | The use of the smart grid for developing intelligent applications is a current trend of great importance. One advantage lies in the possibility of direct monitoring of all devices connected to the electrical network in order to prevent possible malfunctions. Therefore, this paper proposes a method for an automatic detection of the malfunctioning of low-intelligence consumer electrical devices. Malfunctioning means any deviation of a household device from its normal operating schedule. The method is based on a comparison technique, consisting in the correlation between the current power signature of a device and an ideal signature (the standard signature provided by the manufacturer). The first step of this method is to achieve a simplified form of power signature which keeps all the original features. Further, the signal is segmented based on the data provided by an event detection algorithm (values of the first derivatives) and each resulting component is approximated using a regression function. The final step consists of an analysis based on the correlation between the computed regression coefficients and the coefficients of the standard signal. Following this analysis all the differences are classified as a malfunctioning of the analyzed device. |
---|---|
ISSN: | 1582-7445 1844-7600 |
DOI: | 10.4316/AECE.2015.01013 |