Paper Key : IRJ************562
Author: Mujeeb Ali,Dr. Touqeer Ahmed Jumani,Dr. Zohaib Leghari,Engr. Safdar Abro
Date Published: 24 Oct 2023
Abstract
Non-technical losses (NTLs) have been proved to be one of the most challenging concern for electricity distribution companies. Billions of dollars have been projected each year as a result of these criminal actions. There are plenty of causes behind these non-technical losses. One of the biggest reasons is still the use of Traditional time-consuming and inefficient NTL detection methods by power distribution companies around the world, particularly in developing countries. This study aims to address the above-mentioned issue by building an effective energy theft detection model for detecting fraudulent users in a power distribution system. The primary goal of this research is to support distribution system operators in their fight against energy theft. To begin, the proposed computational model Extra-tree classifier Technique is used and number of diverse features which were extracted from monthly consumer consumption data that is obtained from MEPCO to distinguish between honest and fraudulent customers. For many years, electricity theft and energy consumption fraud have been major issues for power distribution companies (PDC). PDCs all over the world are experimenting with various methods for detecting electricity theft. Traditional methods for detecting NTLs, such like on-site inspection and reward and penalty policies, have fallen out of favor in the modern era due to their ineffective and time-consuming methods.Keywords: Non-technical losses (NTLs), Power distribution companies (PDC), Electricity theft.