ENHANCEMENT DETECTION OF ELECTRICITY THEFT UNDER PART XIV OF ELECTRICITY CRIMINAL ACT, 2003 IN DISTRIBUTION SYSTEMS USING RANDOM FOREST FED EXTREME LEARNING MACHINE

Authors

  • Dr. KARTHIKEYAN RAMASAMY

DOI:

https://doi.org/10.52152/801510

Keywords:

Random Forest (RF), Electrical Theft, Support Vector Machine (SVM), Categerical Boosting (Catboost), Extreme Learning Machine (ELM), Confusion Matrix.

Abstract

Day-to-day electricity fraud and thefts in utilities have increased by distribution and consumer levels, and they are also considered non-technical losses. The primary objective of this work is to detect electrical power theft in the distribution system by using a Random Forest (RF) algorithm fed Extreme Learning Machine (ELM), Extereme Gradient Boosting (XGBoost) and Categerical Boosting (CatBoost),  which is an advanced algorithm from aCatboostRF for Support Vector Machine (SVM) classifier. The paper proposes the framework of training a models and testing a models of the data by using information data set to predict the target value. It gives them more efficient when compared to the already used techniques. The two-level data processing approach is proposed in this paper since the data processed by CatboostRF are provided as input to the ELM classifier. The results show that it reduces the deviation error and increases the classification rate accuracy.

Downloads

Published

2025-09-15

Issue

Section

Article

How to Cite

ENHANCEMENT DETECTION OF ELECTRICITY THEFT UNDER PART XIV OF ELECTRICITY CRIMINAL ACT, 2003 IN DISTRIBUTION SYSTEMS USING RANDOM FOREST FED EXTREME LEARNING MACHINE. (2025). Lex Localis - Journal of Local Self-Government, 23(10), 1621-1640. https://doi.org/10.52152/801510