Skip to main navigation Skip to search Skip to main content

Adaptive Control of Converter Parameters Based on DDPG Algorithm

  • School of Electrical Engineering and Automation, Harbin Institute of Technology
  • Harbin Institute of Technology
  • Melent'ev Institute of Power Engineering Systems
  • Irkutsk National Research Technical University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

To improve the control issues of large-scale wind power grid connection, the optimization of control parameters for virtual synchronous generator (VSG) is of vital importance for system stability. This paper proposes an adaptive control strategy for grid-forming converters based on the Deep Deterministic Policy Gradient (DDPG) algorithm to solve the problems of dynamic oscillation of active power and frequency during grid connection, and to enhance the system response speed. Firstly, the mathematical model and characteristics of VSG are analyzed, relevant formulas are derived, root locus curves are plotted, and the ranges and adaptive rules of virtual inertia J and virtual damping coefficient D are obtained. Then, an adaptive parameter controller model is built based on the DDPG algorithm, and the algorithm flow and reward function are detailed. Through simulation comparisons between fixed parameter control and DDPG algorithm optimization control, the results verify the effectiveness of the algorithm in actual operation and providing strong support for the optimization operation of grid-forming converters and wind farms.

Original languageEnglish
Title of host publicationThe Proceedings of 2025 International Conference of Electrical, Electronic and Networked Energy Systems - Volume 1
EditorsLimin Jia, Lei Qi, Zhuangzhuang Liu, Hao Chen, Xianfeng Xu, Baoquan Wei
PublisherSpringer Science and Business Media Deutschland GmbH
Pages237-245
Number of pages9
ISBN (Print)9789819205486
DOIs
StatePublished - 2026
Externally publishedYes
EventInternational Conference of Electrical, Electronic and Networked Energy Systems, EENES 2025 - Hangzhou, China
Duration: 31 Oct 20252 Nov 2025

Publication series

NameLecture Notes in Electrical Engineering
Volume1638 LNEE
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

ConferenceInternational Conference of Electrical, Electronic and Networked Energy Systems, EENES 2025
Country/TerritoryChina
CityHangzhou
Period31/10/252/11/25

Keywords

  • Adaptive control
  • DDPG
  • Deep reinforcement learning
  • Grid-forming converters
  • Virtual synchronous generator

Fingerprint

Dive into the research topics of 'Adaptive Control of Converter Parameters Based on DDPG Algorithm'. Together they form a unique fingerprint.

Cite this