Application of Data Analytics for the Identification of Energy Theft in Electric Power Distribution Systems

• Grant number: 24/06242-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: August 01, 2024
• End date: July 31, 2025
• Field of knowledge: Engineering - Electrical Engineering - Power Systems
• Principal Investigator: Fernanda Caseño Trindade Arioli
• Grantee: Ivan de Sousa Oliveira
• Host Institution: Faculdade de Engenharia Elétrica e de Computação (FEEC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

With the increasing concern about the electrification of the economy, it has become mandatory to ensure greater efficiency in the use of electrical energy transmission and distribution systems. Loss mitigation is a crucial process for distribution utilities in this regard. One common form of non-technical loss in electrical energy distribution systems is electricity theft. This practice is not limited to underdeveloped or developing countries but is also observed in developed countries due to various reasons, such as disguising illicit practices and reducing cryptocurrency mining costs. The annual loss caused by energy theft is around US$25 billion, which is equivalent to the GDP of some African countries. Therefore, to mitigate non-technical losses and consider the increasing availability of data from electrical energy distribution systems, a scientific initiation project aims to analyze real data from a Brazilian utility (installation data, consumption history, and inspection history) to identify rules that may indicate the occurrence of electricity theft. The rules will be based on understanding the problem, data analysis, and pattern recognition. Statistical and clustering techniques will be prioritized for data analysis.