Research Article
Heuristics-Based Detection of Abnormal Energy Consumption
@INPROCEEDINGS{10.1007/978-3-319-94965-9_3, author={Ankur Sial and Amarjeet Singh and Aniket Mahanti and Mingwei Gong}, title={Heuristics-Based Detection of Abnormal Energy Consumption}, proceedings={Smart Grid and Innovative Frontiers in Telecommunications. Third International Conference, SmartGIFT 2018, Auckland, New Zealand, April 23-24, 2018, Proceedings}, proceedings_a={SMARTGIFT}, year={2018}, month={7}, keywords={}, doi={10.1007/978-3-319-94965-9_3} }- Ankur Sial
Amarjeet Singh
Aniket Mahanti
Mingwei Gong
Year: 2018
Heuristics-Based Detection of Abnormal Energy Consumption
SMARTGIFT
Springer
DOI: 10.1007/978-3-319-94965-9_3
Abstract
This paper presents two methods for detecting abnormal electricity consumption by utilizing usage patterns in the vicinity. The methods use contextual and factual information including, energy consumption patterns, nature of supply and category of day to logically group meters and find abnormalities. Using heuristics proposed in the paper, data collected from fifty smart meters deployed inside hostels of IIIT-Delhi were investigated for abnormal electricity consumption. Multiple abnormalities were found and their causes were verified after discussion with campus administrators. Our results show that the proposed heuristics successfully found abnormal energy consumption behavior. Therefore, these methods could be used for real-time abnormality detection. This will result in reducing operating costs by automatically detecting and reporting abnormalities without human intervention.
