• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.8 no.6
• /
• pp.534-543
• /
• 2015

This paper proposes an Energy Theft Traceback Protocol(ETTP) based on Logging and Marking that can trace Energy Theft back in Smart Grid Environment. The ETTP consists of the following three phases. First, it classifies Energy Theft Type into Measurement Rejection and Data Fabrication by generating an Energy Theft Tree. Second, it detects an Energy Theft by using the Energy Theft Tree. Finally, it trace an Energy Theft back by using the Logging Table of a Router and the Marking Information of a Packet. The result of its simulation shows that the Detection Ratio of Energy Theft is estimated at 92% and the Success Ratio of Energy Theft Traceback at 93%. Therefore, the ETTP not only reduces such risk factors as Forgery and Tampering about Billing information but also provides safe and reliable Smart Grid environment.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.5
• /
• pp.119-125
• /
• 2021

As the electricity grid systems has been intelligent with the development of ICT technology, power consumption information of users connected to the grid is available to acquired and analyzed for the power utilities. In this paper, the energy theft problem is solved by feature selection methods, which is emerging as the main cause of economic loss in smart grid. The data preprocessing steps of the proposed system consists of five steps. In the feature selection step, features are selected using analysis of variance and mutual information (MI) based method, which are filtering-based feature selection methods. According to the simulation results, the performance of support vector machine classifier is higher than the case of using all the input features of the input data for the case of the MI based feature selection method.