Deep Reinforcement Learning-based Robust Control with STATCOM for Mitigating Sub- and Super-synchronous Oscillations

Yeunggurl Yoon; Joonhyeok Jang; Xuehan Zhang; Jintae Cho; Sungyun Choi

doi:10.1109/IAS62731.2025.11061527

Deep Reinforcement Learning-based Robust Control with STATCOM for Mitigating Sub- and Super-synchronous Oscillations

Yeunggurl Yoon^*
, Joonhyeok Jang
, Xuehan Zhang
, Jintae Cho
, Sungyun Choi

^*Corresponding author for this work

School of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Regarding resonance and converter-driven stability, sub- and super-synchronous oscillations (SSO and SPO) occur in power systems with high penetration of IBRs. Integrating various IBRs with complex control systems presents challenges in mitigating oscillations across the entire system. Multiple sources, causes, and frequencies contribute to power system oscillations. This paper proposes a deep reinforcement learning (DRL)-based robust oscillation mitigation controller integrated with the STAT-COM controller. A single transmission line system, comprising both IBR and STATCOM, is designed to simulate oscillations and generate cases to train the DRL model. The proposed method is tested on untrained oscillation cases to evaluate the robustness of the controller. As a result, the trained DRL-based controller appropriately operates on steady, transient, and oscillating states.

Original language	English
Title of host publication	2025 IEEE Industry Applications Society Annual Meeting, IAS 2025
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781665457767
DOIs	https://doi.org/10.1109/IAS62731.2025.11061527
Publication status	Published - 2025
Event	2025 IEEE Industry Applications Society Annual Meeting, IAS 2025 - Taipei, Taiwan, Province of China Duration: 2025 Jun 15 → 2025 Jun 20

Publication series

Name	Conference Record - IAS Annual Meeting (IEEE Industry Applications Society)
ISSN (Print)	0197-2618

Conference

Conference	2025 IEEE Industry Applications Society Annual Meeting, IAS 2025
Country/Territory	Taiwan, Province of China
City	Taipei
Period	25/6/15 → 25/6/20

Bibliographical note

Publisher Copyright:
© 2025 IEEE.

Keywords

Deep reinforcement learning
inverter-based resources
sub-synchronous oscillation
super-synchronous oscillation
weak grid

ASJC Scopus subject areas

Control and Systems Engineering
Industrial and Manufacturing Engineering
Electrical and Electronic Engineering

Access to Document

10.1109/IAS62731.2025.11061527

Cite this

Yoon, Y., Jang, J., Zhang, X., Cho, J., & Choi, S. (2025). Deep Reinforcement Learning-based Robust Control with STATCOM for Mitigating Sub- and Super-synchronous Oscillations. In 2025 IEEE Industry Applications Society Annual Meeting, IAS 2025 (Conference Record - IAS Annual Meeting (IEEE Industry Applications Society)). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IAS62731.2025.11061527

Deep Reinforcement Learning-based Robust Control with STATCOM for Mitigating Sub- and Super-synchronous Oscillations. / Yoon, Yeunggurl; Jang, Joonhyeok; Zhang, Xuehan et al.
2025 IEEE Industry Applications Society Annual Meeting, IAS 2025. Institute of Electrical and Electronics Engineers Inc., 2025. (Conference Record - IAS Annual Meeting (IEEE Industry Applications Society)).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Yoon, Y, Jang, J, Zhang, X, Cho, J & Choi, S 2025, Deep Reinforcement Learning-based Robust Control with STATCOM for Mitigating Sub- and Super-synchronous Oscillations. in 2025 IEEE Industry Applications Society Annual Meeting, IAS 2025. Conference Record - IAS Annual Meeting (IEEE Industry Applications Society), Institute of Electrical and Electronics Engineers Inc., 2025 IEEE Industry Applications Society Annual Meeting, IAS 2025, Taipei, Taiwan, Province of China, 25/6/15. https://doi.org/10.1109/IAS62731.2025.11061527

Yoon Y, Jang J, Zhang X, Cho J, Choi S. Deep Reinforcement Learning-based Robust Control with STATCOM for Mitigating Sub- and Super-synchronous Oscillations. In 2025 IEEE Industry Applications Society Annual Meeting, IAS 2025. Institute of Electrical and Electronics Engineers Inc. 2025. (Conference Record - IAS Annual Meeting (IEEE Industry Applications Society)). doi: 10.1109/IAS62731.2025.11061527

Yoon, Yeunggurl ; Jang, Joonhyeok ; Zhang, Xuehan et al. / Deep Reinforcement Learning-based Robust Control with STATCOM for Mitigating Sub- and Super-synchronous Oscillations. 2025 IEEE Industry Applications Society Annual Meeting, IAS 2025. Institute of Electrical and Electronics Engineers Inc., 2025. (Conference Record - IAS Annual Meeting (IEEE Industry Applications Society)).

@inproceedings{07876a1ae8e3487a85939ed569cfb6c6,

title = "Deep Reinforcement Learning-based Robust Control with STATCOM for Mitigating Sub- and Super-synchronous Oscillations",

abstract = "Regarding resonance and converter-driven stability, sub- and super-synchronous oscillations (SSO and SPO) occur in power systems with high penetration of IBRs. Integrating various IBRs with complex control systems presents challenges in mitigating oscillations across the entire system. Multiple sources, causes, and frequencies contribute to power system oscillations. This paper proposes a deep reinforcement learning (DRL)-based robust oscillation mitigation controller integrated with the STAT-COM controller. A single transmission line system, comprising both IBR and STATCOM, is designed to simulate oscillations and generate cases to train the DRL model. The proposed method is tested on untrained oscillation cases to evaluate the robustness of the controller. As a result, the trained DRL-based controller appropriately operates on steady, transient, and oscillating states.",

keywords = "Deep reinforcement learning, inverter-based resources, sub-synchronous oscillation, super-synchronous oscillation, weak grid",

author = "Yeunggurl Yoon and Joonhyeok Jang and Xuehan Zhang and Jintae Cho and Sungyun Choi",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 2025 IEEE Industry Applications Society Annual Meeting, IAS 2025 ; Conference date: 15-06-2025 Through 20-06-2025",

year = "2025",

doi = "10.1109/IAS62731.2025.11061527",

language = "English",

series = "Conference Record - IAS Annual Meeting (IEEE Industry Applications Society)",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2025 IEEE Industry Applications Society Annual Meeting, IAS 2025",

}

TY - GEN

T1 - Deep Reinforcement Learning-based Robust Control with STATCOM for Mitigating Sub- and Super-synchronous Oscillations

AU - Yoon, Yeunggurl

AU - Jang, Joonhyeok

AU - Zhang, Xuehan

AU - Cho, Jintae

AU - Choi, Sungyun

PY - 2025

Y1 - 2025

N2 - Regarding resonance and converter-driven stability, sub- and super-synchronous oscillations (SSO and SPO) occur in power systems with high penetration of IBRs. Integrating various IBRs with complex control systems presents challenges in mitigating oscillations across the entire system. Multiple sources, causes, and frequencies contribute to power system oscillations. This paper proposes a deep reinforcement learning (DRL)-based robust oscillation mitigation controller integrated with the STAT-COM controller. A single transmission line system, comprising both IBR and STATCOM, is designed to simulate oscillations and generate cases to train the DRL model. The proposed method is tested on untrained oscillation cases to evaluate the robustness of the controller. As a result, the trained DRL-based controller appropriately operates on steady, transient, and oscillating states.

AB - Regarding resonance and converter-driven stability, sub- and super-synchronous oscillations (SSO and SPO) occur in power systems with high penetration of IBRs. Integrating various IBRs with complex control systems presents challenges in mitigating oscillations across the entire system. Multiple sources, causes, and frequencies contribute to power system oscillations. This paper proposes a deep reinforcement learning (DRL)-based robust oscillation mitigation controller integrated with the STAT-COM controller. A single transmission line system, comprising both IBR and STATCOM, is designed to simulate oscillations and generate cases to train the DRL model. The proposed method is tested on untrained oscillation cases to evaluate the robustness of the controller. As a result, the trained DRL-based controller appropriately operates on steady, transient, and oscillating states.

KW - Deep reinforcement learning

KW - inverter-based resources

KW - sub-synchronous oscillation

KW - super-synchronous oscillation

KW - weak grid

UR - https://www.scopus.com/pages/publications/105011029627

U2 - 10.1109/IAS62731.2025.11061527

DO - 10.1109/IAS62731.2025.11061527

M3 - Conference contribution

AN - SCOPUS:105011029627

T3 - Conference Record - IAS Annual Meeting (IEEE Industry Applications Society)

BT - 2025 IEEE Industry Applications Society Annual Meeting, IAS 2025

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2025 IEEE Industry Applications Society Annual Meeting, IAS 2025

Y2 - 15 June 2025 through 20 June 2025

ER -

Deep Reinforcement Learning-based Robust Control with STATCOM for Mitigating Sub- and Super-synchronous Oscillations

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this