Data-Driven Coordinated Control of AVR and PSS in Power Systems: A Deep Reinforcement Learning Method

Arman Oshnoei; Omid  Sadeghian; Behnam  Mohammadi-Ivatloo; Frede Blaabjerg; Amjad Anvari-Moghaddam

doi:10.1109/EEEIC/ICPSEurope51590.2021.9584640

Data-Driven Coordinated Control of AVR and PSS in Power Systems: A Deep Reinforcement Learning Method

Arman Oshnoei, Omid Sadeghian , Behnam Mohammadi-Ivatloo, Frede Blaabjerg, Amjad Anvari-Moghaddam^*

^*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

6 Citations (Scopus)

151 Downloads (Pure)

Abstract

In this paper, a strategy based on deep reinforcement learning (DRL) as an intelligent coordinator for power system stabilizer (PSS) and automatic voltage regulator (AVR) in a two-are power grid is proposed. The proposed coordinator is developed to provide accurate online modification of the gains appearing in the structure of PSS and AVR which avoids unfavorable interactions between PSS and AVR under significant changes in the working point and thereby guaranteeing the stability of the power grid. A Markov decision manner is used to formulate the DRL problem and it is solved through a deep deterministic policy gradient approach with an actor-critic framework. Since the intelligent coordinator relies on the expert's science, some scaling coefficients are added to the coordinator body to achieve optimal performance. To confirm the effectiveness of the presented DRL approach, the design is conducted on Kundur's power grid. Simulations illustrate that the proposed DRL-based control can confirm the stability of the system and attain desired dynamic responses.

Original language	English
Title of host publication	2021 IEEE International Conference on Environment and Electrical Engineering : EEEIC 2021
Publisher	IEEE Press
Publication date	2021
Pages	1-6
ISBN (Print)	978-1-6654-3614-4
ISBN (Electronic)	978-1-6654-3613-7
DOIs	https://doi.org/10.1109/EEEIC/ICPSEurope51590.2021.9584640
Publication status	Published - 2021
Event	2021 IEEE International Conference on Environment and Electrical Engineering and 2021 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe) - Bari, Italy Duration: 7 Sept 2021 → 10 Sept 2021

Conference

Conference	2021 IEEE International Conference on Environment and Electrical Engineering and 2021 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe)
Country/Territory	Italy
City	Bari
Period	07/09/2021 → 10/09/2021

Keywords

Power System Stability
Deep Reinforcement Learning
Coordinated Control
Interconnected power systems

Access to Document

10.1109/EEEIC/ICPSEurope51590.2021.9584640

Stability_final_AAM_authors AcceptedAccepted author manuscript, 1.62 MB

AUB Link

Search for the material in Aalborg University Library's search engine

A High-Fidelity PHIL-based Platform for Real-Time Simulation and Testing of Power and Energy Systems
Anvari-Moghaddam, A., Oshnoei, A. & Ribeiro, R. L. D. A.
01/08/2021 → 28/02/2022
Project: Research

File
HeatReFlex: Green and Flexible District Heating/Cooling
Anvari-Moghaddam, A., Guerrero, J. M., Nami, H. & Mohammadiivatloo, B.
Danida Fellowship Centre
01/05/2019 → 30/04/2022
Project: Research

Cite this

Oshnoei, A., Sadeghian , O., Mohammadi-Ivatloo, B., Blaabjerg, F., & Anvari-Moghaddam, A. (2021). Data-Driven Coordinated Control of AVR and PSS in Power Systems: A Deep Reinforcement Learning Method. In 2021 IEEE International Conference on Environment and Electrical Engineering : EEEIC 2021 (pp. 1-6). IEEE Press. https://doi.org/10.1109/EEEIC/ICPSEurope51590.2021.9584640

@inproceedings{4f5a0efe4da8493983d9a07dd2d237a5,

title = "Data-Driven Coordinated Control of AVR and PSS in Power Systems: A Deep Reinforcement Learning Method",

abstract = "In this paper, a strategy based on deep reinforcement learning (DRL) as an intelligent coordinator for power system stabilizer (PSS) and automatic voltage regulator (AVR) in a two-are power grid is proposed. The proposed coordinator is developed to provide accurate online modification of the gains appearing in the structure of PSS and AVR which avoids unfavorable interactions between PSS and AVR under significant changes in the working point and thereby guaranteeing the stability of the power grid. A Markov decision manner is used to formulate the DRL problem and it is solved through a deep deterministic policy gradient approach with an actor-critic framework. Since the intelligent coordinator relies on the expert's science, some scaling coefficients are added to the coordinator body to achieve optimal performance. To confirm the effectiveness of the presented DRL approach, the design is conducted on Kundur's power grid. Simulations illustrate that the proposed DRL-based control can confirm the stability of the system and attain desired dynamic responses.",

keywords = "Power System Stability, Deep Reinforcement Learning, Coordinated Control, Interconnected power systems",

author = "Arman Oshnoei and Omid Sadeghian and Behnam Mohammadi-Ivatloo and Frede Blaabjerg and Amjad Anvari-Moghaddam",

year = "2021",

doi = "10.1109/EEEIC/ICPSEurope51590.2021.9584640",

language = "English",

isbn = "978-1-6654-3614-4",

pages = "1--6",

booktitle = "2021 IEEE International Conference on Environment and Electrical Engineering",

publisher = "IEEE Press",

note = "2021 IEEE International Conference on Environment and Electrical Engineering and 2021 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe) ; Conference date: 07-09-2021 Through 10-09-2021",

}

Oshnoei, A, Sadeghian , O, Mohammadi-Ivatloo, B, Blaabjerg, F & Anvari-Moghaddam, A 2021, Data-Driven Coordinated Control of AVR and PSS in Power Systems: A Deep Reinforcement Learning Method. in 2021 IEEE International Conference on Environment and Electrical Engineering : EEEIC 2021. IEEE Press, pp. 1-6, 2021 IEEE International Conference on Environment and Electrical Engineering and 2021 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe), Bari, Italy, 07/09/2021. https://doi.org/10.1109/EEEIC/ICPSEurope51590.2021.9584640

Data-Driven Coordinated Control of AVR and PSS in Power Systems: A Deep Reinforcement Learning Method. / Oshnoei, Arman; Sadeghian , Omid ; Mohammadi-Ivatloo, Behnam et al.
2021 IEEE International Conference on Environment and Electrical Engineering : EEEIC 2021. IEEE Press, 2021. p. 1-6.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Data-Driven Coordinated Control of AVR and PSS in Power Systems: A Deep Reinforcement Learning Method

AU - Oshnoei, Arman

AU - Sadeghian , Omid

AU - Mohammadi-Ivatloo, Behnam

AU - Blaabjerg, Frede

AU - Anvari-Moghaddam, Amjad

PY - 2021

Y1 - 2021

N2 - In this paper, a strategy based on deep reinforcement learning (DRL) as an intelligent coordinator for power system stabilizer (PSS) and automatic voltage regulator (AVR) in a two-are power grid is proposed. The proposed coordinator is developed to provide accurate online modification of the gains appearing in the structure of PSS and AVR which avoids unfavorable interactions between PSS and AVR under significant changes in the working point and thereby guaranteeing the stability of the power grid. A Markov decision manner is used to formulate the DRL problem and it is solved through a deep deterministic policy gradient approach with an actor-critic framework. Since the intelligent coordinator relies on the expert's science, some scaling coefficients are added to the coordinator body to achieve optimal performance. To confirm the effectiveness of the presented DRL approach, the design is conducted on Kundur's power grid. Simulations illustrate that the proposed DRL-based control can confirm the stability of the system and attain desired dynamic responses.

AB - In this paper, a strategy based on deep reinforcement learning (DRL) as an intelligent coordinator for power system stabilizer (PSS) and automatic voltage regulator (AVR) in a two-are power grid is proposed. The proposed coordinator is developed to provide accurate online modification of the gains appearing in the structure of PSS and AVR which avoids unfavorable interactions between PSS and AVR under significant changes in the working point and thereby guaranteeing the stability of the power grid. A Markov decision manner is used to formulate the DRL problem and it is solved through a deep deterministic policy gradient approach with an actor-critic framework. Since the intelligent coordinator relies on the expert's science, some scaling coefficients are added to the coordinator body to achieve optimal performance. To confirm the effectiveness of the presented DRL approach, the design is conducted on Kundur's power grid. Simulations illustrate that the proposed DRL-based control can confirm the stability of the system and attain desired dynamic responses.

KW - Power System Stability

KW - Deep Reinforcement Learning

KW - Coordinated Control

KW - Interconnected power systems

U2 - 10.1109/EEEIC/ICPSEurope51590.2021.9584640

DO - 10.1109/EEEIC/ICPSEurope51590.2021.9584640

M3 - Article in proceeding

SN - 978-1-6654-3614-4

SP - 1

EP - 6

BT - 2021 IEEE International Conference on Environment and Electrical Engineering

PB - IEEE Press

T2 - 2021 IEEE International Conference on Environment and Electrical Engineering and 2021 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe)

Y2 - 7 September 2021 through 10 September 2021

ER -

Data-Driven Coordinated Control of AVR and PSS in Power Systems: A Deep Reinforcement Learning Method

Abstract

Conference

Keywords

Access to Document

AUB Link

Fingerprint

Projects

A High-Fidelity PHIL-based Platform for Real-Time Simulation and Testing of Power and Energy Systems

HeatReFlex: Green and Flexible District Heating/Cooling

Cite this