EUVIMEDEUVIMED entdecken
Crossref

Design of an LSTM and Bi-LSTM-driven intelligent protection scheme for standalone low-voltage DC microgrid

Anindita Deb, Arvind Kumar Jain

Journal of Renewable and Sustainable Energy · 01.11.2025 · 17 (6)

DOI: 10.1063/5.0290701 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Renewable and Sustainable Energy, online erschienen am 16.12.2025. Verfasst von Anindita Deb und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 52 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Standalone DC microgrids are popular in enhancing energy resilience and reliability in the dynamic energy system landscape. However, protection concerns, such as detection of fault types, identification of fault location, fault characteristics, etc., are major challenges in DC microgrids. Furthermore, the inclusion of power electronic-based DC loads and generators in standalone low-voltage DC microgrid systems (LVDCMG) renders traditional fault detection, classification, and location estimation schemes ineffectual. Therefore, early fault detection, precise fault event response, and fault location estimation are needed to ensure accuracy and reliability of the LVDCMG fault management framework. Thus, the emphasis persists on developing sophisticated tools for DC microgrid fault diagnosis. This paper presents a modern, fast, and reliable data-centric tri-layered deep learning-based framework for fault detection, classification, and location estimation in a standalone LVDCMG programmed through Adam optimization. The framework utilizes long short-term memory (LSTM) and bidirectional LSTM (Bi-LSTM) models for different layers of the application. Primarily, the LSTM layers detect and classify the fault in the system, followed by fault location prediction using a Bi-LSTM network. MATLAB/SIMULINK is used for the proposed method for simulating a 380 V standalone DC microgrid. The outcomes demonstrate the efficacy of the proposed fault diagnosis strategy, showing substantial improvements over conventional techniques. By capturing intricate temporal correlations in the microgrid data, the Bi-LSTM model offers more accurate fault location prediction in the system. Moreover, robust sensitivity analysis to assess its performance under typical and noisy environments has been encompassed in this work, leading the path for more reliable and sustainable electrical systems.

Zitieren

APA

Deb, A., & Jain, A. K. (2025). Design of an LSTM and Bi-LSTM-driven intelligent protection scheme for standalone low-voltage DC microgrid. Journal of Renewable and Sustainable Energy, 17(6). https://doi.org/10.1063/5.0290701

Vancouver

Deb A, Jain AK. Design of an LSTM and Bi-LSTM-driven intelligent protection scheme for standalone low-voltage DC microgrid. Journal of Renewable and Sustainable Energy. 2025;17(6). doi:10.1063/5.0290701

BibTeX
@article{Deb2025,
  author = {Deb, Anindita and Jain, Arvind Kumar},
  title = {Design of an LSTM and Bi-LSTM-driven intelligent protection scheme for standalone low-voltage DC microgrid},
  journal = {Journal of Renewable and Sustainable Energy},
  year = {2025},
  volume = {17},
  number = {6},
  doi = {10.1063/5.0290701}
}
Kostenlos starten →