Comprehensive study of DC microgrids protection : challenges, cutting-edge techniques, machine-learning-driven solutions

Tools

Elmadawy, Mohamed and Shahein, Ahmed and Ghanem, Abdelhady and Abdelaziz, Almoataz Y. and Deng, Fujin and Ayub, Ahmad S. and Mnider, Abdalbaset and Abulanwar, Sayed (2026) Comprehensive study of DC microgrids protection : challenges, cutting-edge techniques, machine-learning-driven solutions. IET Renewable Power Generation, 20 (1). e70258. ISSN 1752-1424 (https://doi.org/10.1049/rpg2.70258)

[thumbnail of Elmadawy-etal-IETRPG-2026-Comprehensive-study-of-DC-microgrids-protection]
Preview
 Text. Filename: Elmadawy-etal-IETRPG-2026-Comprehensive-study-of-DC-microgrids-protection.pdf
Final Published Version
License: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 logo
Download (5MB)| Preview

Abstract

This paper provides a comprehensive examination of the evolving protection challenges within DC microgrids powered by renewable resources and energy storage systems. It begins by delineating the methodological framework of conventional protection, critically assessing schemes based on current, voltage, and impedance to expose their limitations in dynamic and high-resistance fault scenarios. Then, explores cutting-edge, data-driven solutions, highlighting the transformative potential of machine learning and time-frequency transform techniques for achieving superior fault detection, classification, and location through adaptive intelligence. By offering a detailed comparative analysis across key performance indicators, the paper illuminates the trade-offs between speed, cost, selectivity, and reliability inherent in each approach. Besides, identifies persistent research gaps, including the need for standardized guidelines and secure communication networks. Ultimately, the analysis concludes that the path to resilient DC microgrids lies not in a single solution, but in the strategic development of hybrid protection strategies that synergistically combine the robustness of model-based methods with the adaptive intelligence of data-driven algorithms.

ORCID iDs

Elmadawy, Mohamed, Shahein, Ahmed, Ghanem, Abdelhady, Abdelaziz, Almoataz Y., Deng, Fujin, Ayub, Ahmad S. ORCID logoORCID: https://orcid.org/0000-0001-7128-0917, Mnider, Abdalbaset ORCID logoORCID: https://orcid.org/0009-0007-6212-5238 and Abulanwar, Sayed ORCID logoORCID: https://orcid.org/0000-0002-3396-4020;
CORE (COnnecting REpositories)