Electricity Theft Detection Toolbox
This repository provides an open-source toolbox for electricity theft detection, featuring classical classifiers and a data generator. The toolbox was originally hosted on JianShu and has now been moved to GitHub for better accessibility and version control.
Toolbox Overview
The toolbox consists of four main components:
-
Toolbox A collection of .fig files providing a graphical user interface (GUI). A set of .m files containing the core code for data generation and classical classifiers.
-
Raw Data Real household power load profiles from Block 1 of smart meters in the Low Carbon London project.
-
Generated Data The dataset generated from raw data, as used by the authors in their research. Users can directly utilize this dataset.
-
Original Codes A set of .m files containing the original source code. Unlike the GUI-based toolbox, this version is more flexible, allowing users to modify the structure of the classifiers.
How to Download To access the toolbox, select all files and click the Download button.
Video Instructions
For a step-by-step guide on how to use the toolbox, watch the video tutorial here: Bilibili Video(https://www.bilibili.com/video/BV1ZU4y1A7HS?spm_id_from=333.999.0.0)
If you find this toolbox useful, please cite our paper: W. Liao, B. Bak-Jensen, J. R. Pillai, Z. Yang and Y. Wang, "An Open-Source Toolbox with Classical Classifiers for Electricity Theft Detection," 2021 IEEE 2nd China International Youth Conference on Electrical Engineering (CIYCEE), Chengdu, China, 2021, pp. 1-7, doi: 10.1109/CIYCEE53554.2021.9676911.