PowerGym: A Reinforcement Learning Environment for Volt-Var Control in Power Distribution Systems

Fan, Ting-Han; Lee, Xian Yeow; Wang, Yübo

doi:10.48550/arxiv.2109.03970

Cited by 3 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address the voltage collapse due to the fault, OT-sim sends a command to increase the set point of the voltage regulator through the exciter of the generator within a given time. For the voltage control problem, NREL researchers planned to leverage volt-VAR control using an artificial intelligencebased approach with AWS Sagemaker based on similar work implemented by Siemens (Fan, Lee, and Wang 2021). This approach uses a reinforcement learning framework for performing volt-VAR control through controlling the batteries, transformer, and capacitor banks.…”

Section: Implementation Of Voltage Control In Cyber Range: Operationa...mentioning

confidence: 99%

CloudZero Phase 2 Technical Report

Wallace,

Liao,

Rager

et al. 2024

View full text Add to dashboard Cite

Section: Implementation Of Voltage Control In Cyber Range: Operationa...mentioning

confidence: 99%

CloudZero Phase 2 Technical Report

Wallace,

Liao,

Rager

et al. 2024

View full text Add to dashboard Cite

“…Besides, the Gird2Op framework 10 is an open-source environment for training RL agents to operate power networks, which is the testbed for the Learning to Run a Power Network (L2RPN) challenge [157]. Other recently developed RL environments include RLGC [145] for power system control, gymgrid [158] and OMG [159] for microgrid simulation and control, and PowerGym [160] for voltage control in distribution systems, etc. A variety of test systems and test cases are available in these environments.…”

Section: E Numerical Implementationmentioning

confidence: 99%

Reinforcement Learning for Selective Key Applications in Power Systems: Recent Advances and Future Challenges

Chen

Tang

et al. 2022

IEEE Trans. Smart Grid

184

View full text Add to dashboard Cite

With large-scale integration of renewable generation and distributed energy resources, modern power systems are confronted with new operational challenges, such as growing complexity, increasing uncertainty, and aggravating volatility. Meanwhile, more and more data are becoming available owing to the widespread deployment of smart meters, smart sensors, and upgraded communication networks. As a result, data-driven control techniques, especially reinforcement learning (RL), have attracted surging attention in recent years. This paper provides a comprehensive review of various RL techniques and how they can be applied to decision-making and control in power systems. In particular, we select three key applications, i.e., frequency regulation, voltage control, and energy management, as examples to illustrate RL-based models and solutions. We then present the critical issues in the application of RL, i.e., safety, robustness, scalability, and data. Several potential future directions are discussed as well.

show abstract

Soft Actor-Critic With Integer Actions

Fan¹,

Wang²

2021

Preprint

View full text Add to dashboard Cite

Reinforcement learning is well-studied under discrete actions. Integer actions setting is popular in the industry yet still challenging due to its high dimensionality. To this end, we study reinforcement learning under integer actions by incorporating the Soft Actor-Critic (SAC) algorithm with an integer reparameterization. Our key observation for integer actions is that their discrete structure can be simplified using their comparability property. Hence, the proposed integer reparameterization does not need one-hot encoding and is of low dimensionality. Experiments show that the proposed SAC under integer actions is as good as the continuous action version on robot control tasks and outperforms Proximal Policy Optimization on power distribution systems control tasks.

show abstract

PowerGym: A Reinforcement Learning Environment for Volt-Var Control in Power Distribution Systems

Cited by 3 publications

References 22 publications

CloudZero Phase 2 Technical Report

CloudZero Phase 2 Technical Report

Reinforcement Learning for Selective Key Applications in Power Systems: Recent Advances and Future Challenges

Soft Actor-Critic With Integer Actions

Contact Info

Product

Resources

About