SOFTWARE SOLUTION FOR THE DETECTION OF ELECTRICAL ENERGY CONSUMPTION ANOMALIES IN ML .NET AND PYTHON
DOI:
https://doi.org/10.24867/23BE26GrbicKeywords:
Python, Anomaly detection, Machine learning, ML .NETAbstract
Data from the Smart Grid system can be analyzed to detect abnormal phenomena in various areas such as cyber security, theft detection, failure detection, etc. Machine learning algorithms can be used as a tool for analyzing and processing raw data coming from real-time systems. This paper describes, analyzes and presents the results of the implemented software solution that detects anomalies based on data containing electricity consumption over algorithms in ML .NET and Python.
References
[1] https://bif.rs/2021/12/ukupni-distributivni-gubici-eps-a-su-medju-najvecim-u-evropi/ (pristupljeno u martu 2023.)
[2] Raja Masood Larik, Mohd Wazir Mustafa, Sajid Hussain Qazi, „Smart Grid Technologies in Power System“s, University of Technology Malaysia, Malaysia
[3] https://www.techopedia.com/definition/692/smart-grid (pristupljeno u martu 2023.)
[4] Srdjan Milošević, Emil Naumovski, „Sistem za daljinsko očitavanje i upravljanje potrošnjom u PD „Elektrodistribucija Beograd“”, Zbornik Međunarodnog kongresa o KGH, [S.l.], v. 44, n. 1, p. 1-6, oct. 2017
[5] https://anyline.com/news/detect-non-technical-losses-energy-utility (pristupljeno u martu 2023.)
[6] Bishop, C. M., „Pattern Recognition and Machine Learning, Springer“
[7] Nenad Rakić, “Primena metoda mašinskog učenja za rangiranje individualnih sposobnosti”, Prirodni Matematički Fakultet, Univerzitet u Novom Sadu, 2020,
[8] as. ms Vladimir Jocović, as. ms Adrian Milaković, “Inteligentni sistemi”, Elektrotehnički fakultet, Univerzitet u Beogradu
[9] https://blog.paperspace.com/anomaly-detection-isolation-forest/ (pristupljeno u martu 2023.)
[10] Kumar Reddy Shabad, Abdulmueen Alrshide, „Anomaly Detection in Smart Grids using Machine Learning, Miami, Florida, USA, 2021.
[11] https://learn.microsoft.com/en-us/dotnet/api/microsoft.ml.timeseriescatalog.detectspikebyssa?view=ml-dotnet (pristupljeno u martu 2023.)
[12]https://en.wikipedia.org/wiki/Independent_and_identically_distributed_random_variables (pristupljeno u martu 2023.)
[13] https://towardsdatascience.com/time-series-anomaly-detection-b10fdb542974 (pristupljeno u martu 2023.)
[14] Vladimir Rokhlin, Arthur Szlam, Mark Tygert, „A randomized algorithm for principal component analysis“, Jul 2009.
[15] https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.IsolationForest.html (pristupljeno u martu 2023.)
[2] Raja Masood Larik, Mohd Wazir Mustafa, Sajid Hussain Qazi, „Smart Grid Technologies in Power System“s, University of Technology Malaysia, Malaysia
[3] https://www.techopedia.com/definition/692/smart-grid (pristupljeno u martu 2023.)
[4] Srdjan Milošević, Emil Naumovski, „Sistem za daljinsko očitavanje i upravljanje potrošnjom u PD „Elektrodistribucija Beograd“”, Zbornik Međunarodnog kongresa o KGH, [S.l.], v. 44, n. 1, p. 1-6, oct. 2017
[5] https://anyline.com/news/detect-non-technical-losses-energy-utility (pristupljeno u martu 2023.)
[6] Bishop, C. M., „Pattern Recognition and Machine Learning, Springer“
[7] Nenad Rakić, “Primena metoda mašinskog učenja za rangiranje individualnih sposobnosti”, Prirodni Matematički Fakultet, Univerzitet u Novom Sadu, 2020,
[8] as. ms Vladimir Jocović, as. ms Adrian Milaković, “Inteligentni sistemi”, Elektrotehnički fakultet, Univerzitet u Beogradu
[9] https://blog.paperspace.com/anomaly-detection-isolation-forest/ (pristupljeno u martu 2023.)
[10] Kumar Reddy Shabad, Abdulmueen Alrshide, „Anomaly Detection in Smart Grids using Machine Learning, Miami, Florida, USA, 2021.
[11] https://learn.microsoft.com/en-us/dotnet/api/microsoft.ml.timeseriescatalog.detectspikebyssa?view=ml-dotnet (pristupljeno u martu 2023.)
[12]https://en.wikipedia.org/wiki/Independent_and_identically_distributed_random_variables (pristupljeno u martu 2023.)
[13] https://towardsdatascience.com/time-series-anomaly-detection-b10fdb542974 (pristupljeno u martu 2023.)
[14] Vladimir Rokhlin, Arthur Szlam, Mark Tygert, „A randomized algorithm for principal component analysis“, Jul 2009.
[15] https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.IsolationForest.html (pristupljeno u martu 2023.)
Downloads
Published
2023-07-08
Issue
Section
Electrotechnical and Computer Engineering
