Novel efficient deep reinforcement learning-based load frequency control for isolated microgrid

This study introduces a Learning-based Load Frequency Control (LB-LFC) approach to manage the challenges posed by renewable energy’s intermittency in microgrids, which often causes load disturbances, frequency fluctuations, and higher generation costs. The LB-LFC method employs reinforcement learnin...

Full description

Saved in:
Bibliographic Details
Main Authors: Xin Shen, Yijing Zhang, Jiahao Li, Yitao Zhao, Jianlin Tang, Bin Qian, Xiaoming Lin
Format: Article
Language:English
Published: AIP Publishing LLC 2025-02-01
Series:AIP Advances
Online Access:http://dx.doi.org/10.1063/5.0240774
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study introduces a Learning-based Load Frequency Control (LB-LFC) approach to manage the challenges posed by renewable energy’s intermittency in microgrids, which often causes load disturbances, frequency fluctuations, and higher generation costs. The LB-LFC method employs reinforcement learning to balance generation costs and frequency stability effectively. In addition, a novel sort replay actor critic technique is proposed, leveraging the deep deterministic policy gradient algorithm and sort experience replay to enhance control efficiency and robustness. This dual-objective control strategy not only improves frequency management but also aims to reduce generation expenses. The effectiveness of this approach is validated through simulations on the isolated microgrid load frequency control model of China Southern Grid.
ISSN:2158-3226