Skip to main navigation Skip to search Skip to main content

Deep learning-based predictive control for voltage enhancement and power management in DC MG with hybrid energy storage systems

Research output: Contribution to journal › Article › peer-review

12 Scopus citations

Abstract

The increasing integration of flexible loads, load uncertainty, and fluctuating renewable energy sources (RES) in DC microgrids (MGs) challenges conventional control strategies in maintaining power balance, thereby affecting power quality and necessitating increased investment in hybrid energy storage systems (HESS). This paper proposes a novel Long Short-Term Memory (LSTM) deep learning-based predictive control strategy to enhance DC bus voltage regulation and optimize power management in DC MGs. The proposed method developed a predictive control strategy to manage HESS and optimize power allocation. By separating high-frequency signals, which are assigned to the supercapacitor (SC), from low-frequency signals, which are assigned to the Battery Energy Storage System (BESS), the strategy enhances system performance. It effectively addresses rapid and uncertain load variations as well as power fluctuations, ensuring efficient energy management and stability. The primary purpose of the proposed LSTM strategy is to prevent SC violations by keeping its state of charge (SoC) within the specified range while maintaining the SC's uninterrupted functionality. The effectiveness of the proposed LSTM control method is investigated under different test scenarios, incorporating critical system uncertainties and disturbances. The results demonstrate the effectiveness of the proposed method in enhancing the transient and dynamic response of the system, which leads to enhanced voltage regulation and better power quality of the DC MGs.

Original languageEnglish
Article number118416
JournalJournal of Energy Storage
Volume136
DOIs
StatePublished - 15 Nov 2025

Bibliographical note

Publisher Copyright:
© 2025

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 7 - Affordable and Clean Energy
    SDG 7 Affordable and Clean Energy

Keywords

  • Energy management control
  • Filter-based (FB) control
  • Hybrid energy storage system
  • LSTM
  • Microgrids
  • Renewable energy fluctuations

ASJC Scopus subject areas

  • Renewable Energy, Sustainability and the Environment
  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Deep learning-based predictive control for voltage enhancement and power management in DC MG with hybrid energy storage systems'. Together they form a unique fingerprint.

Cite this