Integrating Digital Twin and Machine Learning for Cybersecurity in Power Electronics-Dominated Grids

View at publisher Summarise with VIRAT
Authors: Kanchan Yadav
Cited by: 0
References: 23

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The industrial city of Manchester is one of the most popular in England which attracts migrants of numerous different countries. In this paper, the Digital Twin based Cyber Security and Machine Learning (ML) System in Google Enterprise Design Grids (EDES) is explain to be one of the key points in that it provides the virtual replicas that exist behind the apparent picture of a physical grid to be in operation. It allows constant surveillance, gathering of data and identification of anomalies. The proposed system has ML algorithms become members of DTs that will learn how a grid should be functioning properly to be able to report trouble points or simply plain attacks made to impair information systems. The work consists of: generation of strong ML models, including Random Forest and Long Short-Term Memory networks of real-time computerassisted threat detection Development of advanced data preprocessing methods, including class imbalance and feature selection, are introduced in the system to achieve substantial improvement in detection performance with data combined DT enhancement. The RF model has been found to have an approximately$\mathbf{1 0}-\mathbf{1 5 {\%}}$increase in accuracy and$\mathbf{F 1}$-score compared with its standalone implementation when DT-aided. To conclude, integrating DT into the same system as ML can bring transformative benefits for safeguarding critical energy infrastructure. Future work will be critical for scaling up to higher-dimensional grids and adding adaptive learning models that handle growing PEDGs as their position evolves.

Fields of Study & Subjects

Smart Grid Security and Resilience Network Security and Intrusion Detection Electricity Theft Detection Techniques

Publication details

Journal / source
International Conference on Power Electronics & IoT Applications in Renewable Energy and its Control (PARC)
Pages
289–293
Publication date
20 February 2026
Language
EN