ANALYSIS OF METHODS FOR DETECTING ANOMALIES IN ELECTRICITY CONSUMPTION DATA

Authors

  • Mykyta Bratyshchenko
  • Tetiana Filimonchuk
  • Halyna Maistrenko
  • Vitalii Sitnikov

DOI:

https://doi.org/10.26906/SUNZ.2024.1.045

Keywords:

anomaly detection, energy consumption, statistical data, machine learning, energy efficiency, clustering, methods

Abstract

Topicality. Modern people are constantly looking for new ways to use energy to improve their lives, so the demand for it is growing. In most cases, it is difficult for companies and industries to control all their devices at the same time, which can lead to a loss of electricity at any time. As a result, operating costs will be higher than necessary. In addition, the loss of power contributes to global warming through the release of carbon when energy is generated by burning coal, gas, and oil. Thus, solutions are needed to address these issues. The purpose of this work is to analyse existing methods for detecting anomalies in data to solve the problem of excessive electricity consumption and to warn of critical values in the indicators of electricity consumed by various devices and electrical equipment. The object of the study is the process of detecting atypical values or significant deviations in electricity consumption by such parameters as voltage, current strength and frequency, power. The subject of the study is models and methods for detecting anomalies in data. Results. After a thorough analysis of each of the above anomaly detection methods, new opportunities for solving the energy consumption problem open up. For example: combining several methods into one; developing a machine learning model based on one or more methods, training on test data and, in the future, processing real energy consumption data to identify atypical values, with the ability to record the date and time of anomalies, and build various graphs based on this information. Conclusion. The anomaly detection methods discussed here can prevent high electricity consumption to achieve energy savings, remind users to identify faulty electrical appliances or change incorrect electricity consumption patterns, reduce users' energy costs, and promote awareness of electricity safety.

Downloads

References

Berardi U. (2015), “Building energy consumption in US, EU, and BRIC countries”, Department of Architectural Science, Faculty of Engineering and Architectural Science, № 118, P. 128-136.

Jia Yan Lim, Wooi-Nee Tan, Yi-Fei Tan (2022), “Anomalous energy consumption detection using a Naïve Bayes approach”, Faculty of Engineering, Multimedia University, Cyberjaya, Selangor, 63100, Malaysia, № 1, P. 4.

Zhang J., Zhang H., Ding S., Znang X. (2021), “Power Consumption Predicting and Anomaly Detection Based on Transformer and K-Means”, College of Mathematics and Inf. Technology, Hebei University. Vol. 9, Article № 779587, P. 3-7.

Longji Feng, Shu Xu, Linghao Zhang, Jing Wu, Jidong Zhang, Chengbo Chu, Zhenyu Wang and Haoyang Shi (2020), “Anomaly detection for electricity consumption in cloud computing: framework, methods, applications, and challenges”, Wireless Com Network, № 194, P. 2-10, doi: https://doi.org/10.1186/s13638-020-01807-0

Liu X., Iftikhar N., Nielsen P.S., Heller A. (2016), “Online Anomaly Energy Consumption Detection Using Lambda Architecture”, Big Data Analytics and Knowledge Discovery, vol 9829, P. 193-209, doi: https://doi.org/10.1007/978-3-319-43946-4_13

Yasirli Amril, Amanda Lailatul Fadhilah, Fatmawati, Novi Setianil, Septia Ranil (2016), “Analysis Clustering of Electricity Usage Profile Using K-Means Algorithm”, IOP Conf. Series. № 105, P.2-7, doi: 10.1088/1757-899X/105/1/012020

Lamrini B., Gjini A., Daudin S., Armando F., Pratmarty P., Travé-Massuyès L. (2018) “Anomaly Detection Using Similarity-based One-Class SVM for Network Traffic Characterization”, Un-té de Toulouse, CNRS, Toulouse, France., № 1, P. 2-4.

Auskalnis J., Paulauskas N., Baskys A. (2018), “Application of Local Outlier Factor Algorithm to Detect Anomalies in Computer Network”, Elektronika I Elektrotechnika, 24(3), P. 96-99, doi: https://doi.org/10.5755/j01.eie.24.3.20972

Nasraoui O., Leon E., Krishnapuram R. (2005), “Unsupervised Niche Clustering: Discovering an Unknown Number of Clusters in Noisy Data Sets”., Evolutionary Computation in Data Mining, Springer Berlin Heidelberg. Volume 1, P. 2-30.

Lizabeth Leon., Olfa Nasraoui., Jonatan Gomez. (2007), “Anomaly detection based on unsupervised niche clustering with application to network intrusion detection”, Proc. of the IEEE Conference on Evolutionary Computation. Volume 1, P. 502.

Chandola V., Banerjee A., Kumar V. (2009), “Anomaly detection: A survey”, ACM Computing Surveys, № 41, P. 1-58.

Melnyk I., Matthews B., Valizadegan H., Banerjee A., Oza N. (2016), “Vector Autoregressive Model-Based Anomaly Detection in Aviation Systems”, JAIS Aerospace Information Systems., 13, P. 161-173.

Published

2024-02-09

Most read articles by the same author(s)