Abstract:

The upgrading of smart grids plays a crucial role in electricity theft detection as they generate massive data which includes customer consumption data. Electricity theft is a global problem that negatively impacting on utility companies and also electricity users. it impairs utility companies' ability to grow economically, creates electric risks, and has an effect on consumers' high energy costs. Power loss in the transmission and distribution of electricity is an important problem faced by power utility companies from all over the world. Power losses are mainly categorized into technical losses and non-technical losses. Technical loss is related to electricity transmission, which is caused by internal scenes in power system components such as linear transmission; NTL is calculated as the difference between total losses and TLs, which is generally done by electricity theft. Basically, electricity theft mostly happens through physical attacks like eavesdropping, breaking the meter, or tampering with the meter reading.

Description:

SVM, Convolutional Neural Networks, Random Forest, theft, smart grid