Loading...

Developing a New Algorithm for Detecting Electricity Theft in Crypto-Currency Miners

Bagheri, Mohsen | 2022

152 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 55593 (46)
  4. University: Sharif University of Technology
  5. Department: Energy Engineering
  6. Advisor(s): Moeini Aghtaei, Moein
  7. Abstract:
  8. Cryptocurrency miners by solving complex mathematical calculations are responsible for verifying the transactions made in the blockchain network as well as maintaining its security, and as a reward for these activities, they receive bitcoins from the network. The devices used to mine cryptocurrency in order to perform the aforementioned calculations need high electricity consumption, so that the main cost of mining is related to its electricity consumption. For this reason, the ever-increasing development of the blockchain network, as well as the significant growth of the value of Bitcoin, has increased the number of cryptocurrency miners, especially in countries with low electricity costs. The sudden increase in electricity demand in these countries has brought many challenges, and it is very important to identify miners in order to control their consumption and also prevent illegal cases.This research deals with the recognition of electricity consumers who are mining cryptocurrency in the network and introduces the group that operates without permission as electricity thieves. In order to identify according to the electricity consumption information of the subscribers, first those consumers who have different consumption behavior than others based on the outlier data detection model and also at the same time the most similarity and behavioral correlation with the changes in electricity consumption in the blockchain network , they are marked as cryptocurrency miners. Then, by having the labeled dataset based on miners, using machine learning algorithms, a model has been trained to identify cryptocurrency miners in the network.
  9. Keywords:
  10. Machine Learning ; Cryptocurrency ; Blockchain ; Cryptocurrency Miners Identification ; Electricity Theft Detection ; Labeling Unsupervised Dataset

 Digital Object List

 Bookmark

No TOC