Skip to main content

ORIGINAL RESEARCH article

Front. Energy Res.
Sec. Smart Grids
Volume 12 - 2024 | doi: 10.3389/fenrg.2024.1517861
This article is part of the Research Topic Optimal Scheduling of Demand Response Resources In Energy Markets For High Trading Revenue and Low Carbon Emissions View all 30 articles

Learning-Driven Load Frequency Control for Islanded Microgrid Using Graph Networks-based Deep Reinforcement Learning

Provisionally accepted
Wangyong Guo Wangyong Guo *Hongwei Du Hongwei Du Tao Han Tao Han Shuang Li Shuang Li Chao Lu Chao Lu Xiaoming Huang Xiaoming Huang
  • Nari Group Corporation State Grid Electric Power Research Institute, Nanjing, China

The final, formatted version of the article will be published soon.

    As the complexity of microgrid systems, the randomness of load disturbances, and the data dimensionality increase, traditional load frequency control methods for microgrids are no longer capable of handling such highly complex and nonlinear control systems. This can result in This can result in significant frequency fluctuations and oscillations, potentially leading to blackouts in microgrids. To address the random power disturbances introduced by a large amount of renewable energy, this paper proposes a Learning-Driven Load Frequency Control (LD-LFC) method. Additionally, a Graph Convolution Neural Networks -Proximal Policy Optimization (GCNN -PPO) algorithm is introduced, which enhances the random power disturbances introduced by a large amount of renewable energy. algorithm is introduced, which enhances the perception ability of the reinforcement learning agent regarding grid state data by embedding a graph convolutional network. The effectiveness of this approach is validated through simulations on the isolated microgrid Load Frequency Control (LFC) model of China Southern Grid (CSG).

    Keywords: Load frequency control, Graph Convolution Neural Networks, Isolated microgrid, multi-objective optimization, Proximal policy optimization

    Received: 27 Oct 2024; Accepted: 28 Nov 2024.

    Copyright: © 2024 Guo, Du, Han, Li, Lu and Huang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence: Wangyong Guo, Nari Group Corporation State Grid Electric Power Research Institute, Nanjing, China

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.