Main Article Content
Abstract
Energy is one of the most vital sources, in which electrical energy is mandatory. The government is taking a lot of steps and investing huge money in electricity generation. Government has suffered economic losses in recent years owing to energy theft. Theft of electricity occurs when a person taps electricity lines, tampers with electric meters or transformers or uses a device that interferes with reading or damages equipment such as electric meters or uses electricity for purposes other than permitted. If the consumers are on the same power line where someone is trying to steal electricity, the consumer should also pay the cost of the theft. Theft of electricity gives the paying customers less power, less efficiency and less price. Theft of energy is a criminal offence and punishable as a fine. Every year $96 trillion worldwide is lost due to energy theft. The main purpose of this paper is to design a system to detect any electricity-related theft, to resolve government economic losses. This framework suggests a method for detecting suspicious customers using the consumption pattern of the customer power. In this project, machine learning algorithms such as the k-means algorithm and ANN are used. Customer trustworthiness is checked and is selected to detect fraud system. And to avoid the inconvenience of users due to theft, bill regardless of the use of electricity due to theft. The illegal operation in the use of electricity shall be reported to the electricity board office automatically.