Electricity Theft Detection Using Deep Neural Network

Authors: Prof.H.R.Agashe, Gite Swarup Sharad, Bhosale Vipul Vijay, Talekar Sonali Rajendra, Nimse Aishwarya Ratan

Country: India

Full-text Research PDF File:   View   |   Download

Abstract: Electricity theft poses a significant challenge to utility companies worldwide, lead ing to revenue losses and compromised service quality. Traditional methods of de tecting theft often fall short in accuracy and efficiency. In this study, we propose a novel approach to electricity theft detection utilizing machine learning techniques. We start by collecting a dataset of one year’s electricity usage by customers, sourced from Kaggle. To enhance the accuracy of our model, we adopt a unique preprocessing step where we convert the dataset into monthly records. This conversion facilitates the cre ation of a more granular and insightful dataset, allowing our system to capture subtle variations in usage patterns over time.Our proposed system operates as follows: when a new user logs into the platform and inputs parameters such as their last month’s bill, our system employs machine learning algorithms to analyze the data and determine whether the customer has experienced electricity theft. By leveraging the month wise dataset, our system can effectively distinguish between legitimate fluctuations in usage and anomalous patterns indicative of theft.Through rigorous experimentation and evaluation on real-world datasets, we demonstrate the efficacy of our approach in accurately detecting electricity theft. Our system not only improves the detection accuracy but also enhances the overall efficiency of theft identification processes for utility companies. We believe that our proposed methodology holds great promise in combating electricity theft and ensuring fair distribution of resources in the energy sector.

Keywords: Electricity theft detection- Machine learning- Monthly data con version- Dataset preprocessing- Utility companies- Anomalous patterns- Detection accuracy- Efficiency- Energy sector.

Paper Id: 230625

Published On: 2024-05-03

Published In: Volume 12, Issue 3, May-June 2024

Share this