A method for the detection of electricity theft behavior based on Xavier weight initialization

Jiaqi Liang

doi:10.1117/12.2685710

15 August 2023 A method for the detection of electricity theft behavior based on Xavier weight initialization

Jiaqi Liang

Proceedings Volume 12719, Second International Conference on Electronic Information Technology (EIT 2023); 1271934 (2023) https://doi.org/10.1117/12.2685710
Event: Second International Conference on Electronic Information Technology (EIT 2023), 2023, Wuhan, China

Abstract

With the advancement of machine learning technology, intelligent detection of electricity theft has become a more widely used technical means. The detection of electricity theft behavior includes data collection, data preprocessing, establishment of electricity theft model and online detection process of electricity theft, in which the establishment of the electricity theft model is modeled by neural network, and the random initialization method and Xavier weight initialization method are used in the modeling process. Experiments show that the Xavier weight initialization method has a good effect on error histogram and network performance, and the mean squared normalization error of 0.0686 is lower than that of the stochastic optimization method.

Citation Download Citation

Jiaqi Liang "A method for the detection of electricity theft behavior based on Xavier weight initialization", Proc. SPIE 12719, Second International Conference on Electronic Information Technology (EIT 2023), 1271934 (15 August 2023); https://doi.org/10.1117/12.2685710

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available