Adaptive Control of VSG Inertia Damping Based on MADDPG
As renewable energy sources become more integrated into the power grid, traditional virtual synchronous generator (VSG) control strategies have become inadequate for the current low-damping, low-inertia power systems. Therefore, this paper proposes a VSG inertia and damping adaptive control method b...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/17/24/6421 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850059426997731328 |
|---|---|
| author | Demu Zhang Jing Zhang Yu He Tao Shen Xingyan Liu |
| author_facet | Demu Zhang Jing Zhang Yu He Tao Shen Xingyan Liu |
| author_sort | Demu Zhang |
| collection | DOAJ |
| description | As renewable energy sources become more integrated into the power grid, traditional virtual synchronous generator (VSG) control strategies have become inadequate for the current low-damping, low-inertia power systems. Therefore, this paper proposes a VSG inertia and damping adaptive control method based on multi-agent deep deterministic policy gradient (MADDPG). The paper first introduces the working principles of virtual synchronous generators and establishes a corresponding VSG model. Based on this model, the influence of variations in virtual inertia (<i>J</i>) and damping (<i>D</i>) coefficients on fluctuations in active power output is examined, defining the action space for <i>J</i> and <i>D</i>. The proposed method is mainly divided into two phases: “centralized training and decentralized execution”. In the centralized training phase, each agent’s critic network shares global observation and action information to guide the actor network in policy optimization. In the decentralized execution phase, agents observe frequency deviations and the rate at which angular frequency changes, using reinforcement learning algorithms to adjust the virtual inertia <i>J</i> and damping coefficient <i>D</i> in real time. Finally, the effectiveness of the proposed MADDPG control strategy is validated through comparison with adaptive control and DDPG control methods. |
| format | Article |
| id | doaj-art-eda079f332f247c78313e4ab1b830f5f |
| institution | DOAJ |
| issn | 1996-1073 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj-art-eda079f332f247c78313e4ab1b830f5f2025-08-20T02:50:53ZengMDPI AGEnergies1996-10732024-12-011724642110.3390/en17246421Adaptive Control of VSG Inertia Damping Based on MADDPGDemu Zhang0Jing Zhang1Yu He2Tao Shen3Xingyan Liu4College of Electrical Engineering, Guizhou University, Guiyang 550025, ChinaCollege of Electrical Engineering, Guizhou University, Guiyang 550025, ChinaCollege of Electrical Engineering, Guizhou University, Guiyang 550025, ChinaCollege of Electrical Engineering, Guizhou University, Guiyang 550025, ChinaPower Grid Planning and Research Center of Guizhou Power Grid Co., Ltd., Guiyang 550002, ChinaAs renewable energy sources become more integrated into the power grid, traditional virtual synchronous generator (VSG) control strategies have become inadequate for the current low-damping, low-inertia power systems. Therefore, this paper proposes a VSG inertia and damping adaptive control method based on multi-agent deep deterministic policy gradient (MADDPG). The paper first introduces the working principles of virtual synchronous generators and establishes a corresponding VSG model. Based on this model, the influence of variations in virtual inertia (<i>J</i>) and damping (<i>D</i>) coefficients on fluctuations in active power output is examined, defining the action space for <i>J</i> and <i>D</i>. The proposed method is mainly divided into two phases: “centralized training and decentralized execution”. In the centralized training phase, each agent’s critic network shares global observation and action information to guide the actor network in policy optimization. In the decentralized execution phase, agents observe frequency deviations and the rate at which angular frequency changes, using reinforcement learning algorithms to adjust the virtual inertia <i>J</i> and damping coefficient <i>D</i> in real time. Finally, the effectiveness of the proposed MADDPG control strategy is validated through comparison with adaptive control and DDPG control methods.https://www.mdpi.com/1996-1073/17/24/6421VSGmulti-agentdeep deterministic policy gradientfrequency control |
| spellingShingle | Demu Zhang Jing Zhang Yu He Tao Shen Xingyan Liu Adaptive Control of VSG Inertia Damping Based on MADDPG Energies VSG multi-agent deep deterministic policy gradient frequency control |
| title | Adaptive Control of VSG Inertia Damping Based on MADDPG |
| title_full | Adaptive Control of VSG Inertia Damping Based on MADDPG |
| title_fullStr | Adaptive Control of VSG Inertia Damping Based on MADDPG |
| title_full_unstemmed | Adaptive Control of VSG Inertia Damping Based on MADDPG |
| title_short | Adaptive Control of VSG Inertia Damping Based on MADDPG |
| title_sort | adaptive control of vsg inertia damping based on maddpg |
| topic | VSG multi-agent deep deterministic policy gradient frequency control |
| url | https://www.mdpi.com/1996-1073/17/24/6421 |
| work_keys_str_mv | AT demuzhang adaptivecontrolofvsginertiadampingbasedonmaddpg AT jingzhang adaptivecontrolofvsginertiadampingbasedonmaddpg AT yuhe adaptivecontrolofvsginertiadampingbasedonmaddpg AT taoshen adaptivecontrolofvsginertiadampingbasedonmaddpg AT xingyanliu adaptivecontrolofvsginertiadampingbasedonmaddpg |