Efficient Electricity Theft Detection Using Machine Learning Algorithms

Hrishikesh Mohan Dabir, Aditya Suresh Kadam, Gaurav Hadge, Ayushman Singh Rathore, Prof. Shubhangi Ingale1

1

Publication Date: 2020/04/17

Abstract: Electricity theft is one of the major problems of electric utilities. Such electricity theft produce financial loss to the utility companies. It is not possible to inspect manually such theft in large amount of data. For detecting electricity theft introduces a gradient boosting theft detector (GBTD) which utilizes three gradient based classifiers also known as (GBCs) which can be boosted that are extreme gradient boosting (XGBoost), categorical boosting (Cat Boost), and method as (LightGBM).XGBoost is one machine learning algorithm which gives high accuracy in less time.In this we apply preprocessing on smart meter data then does feature selection.Various application of the given GBTD is for electricity theft detection by reducing time taken to generate results of the GBTD model which detects nontechnical loss (NTL) detection.

Keywords: Artificial Intelligence(AI), Artifical Neural Network(ANN), XGBoost, CatBoost, Light GBM, electricity theft detection, gradient boosting.

DOI: No DOI Available

PDF: https://ijirst.demo4.arinfotech.co/assets/upload/files/IJISRT19DEC371_(1).pdf

REFERENCES

No References Available