Revenue and Reliability Protection for Energy Suppliers From Energy Theft

Dipu Sarkar, Prasanta K. Ghosh, Ch Sekhar Gujjarlapudi

Research output: Chapter in Book/Entry/PoemConference contribution

Abstract

Energy theft poses technical and social challenges for utilities and legitimate customers. In recent years, numerous machine learning (ML) concepts have been used to enhance power system operations in support of energy sector transitions towards digitization. The objective of the work is to identify suitable approaches to counteract various energy theft methods. In this collaborative work we have applied ML models with the market demand trends to find the most suitable approach(es) for identifying energy stealing. The work begins with some relevant results using ensemble ML models for energy theft detection. We then introduce the Naive Bayes model, which has been used for handling classification problems through supervised learning. This model is used to develop a methodology for identifying energy theft based on customers' consumption patterns. Results were compared using false-positive and detection rates for assessing their effectiveness.

Original languageEnglish (US)
Title of host publication2024 IEEE 12th International Conference on Smart Energy Grid Engineering, SEGE 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages156-159
Number of pages4
ISBN (Electronic)9798350377378
DOIs
StatePublished - 2024
Event12th IEEE International Conference on Smart Energy Grid Engineering, SEGE 2024 - Oshawa, Canada
Duration: Aug 18 2024Aug 20 2024

Publication series

Name2024 IEEE 12th International Conference on Smart Energy Grid Engineering, SEGE 2024

Conference

Conference12th IEEE International Conference on Smart Energy Grid Engineering, SEGE 2024
Country/TerritoryCanada
CityOshawa
Period8/18/248/20/24

Keywords

  • Energy theft
  • machine learning
  • naive bayes

ASJC Scopus subject areas

  • Artificial Intelligence
  • Energy Engineering and Power Technology
  • Renewable Energy, Sustainability and the Environment
  • Automotive Engineering
  • Electrical and Electronic Engineering
  • Mechanical Engineering
  • Control and Optimization

Fingerprint

Dive into the research topics of 'Revenue and Reliability Protection for Energy Suppliers From Energy Theft'. Together they form a unique fingerprint.

Cite this