Deep-Reinforcement-Learning-Based Autonomous Voltage Control for Power Grid Operations

By Jiajun Duan, Di Shi, R. Diao, Haifeng Li, Bei Zhang, Desong Bian, Zhehan Yi. IEEE Transactions on Power Systems, 2020

362 citations recorded by the source corpus.

In this letter, a novel autonomous control framework “Grid Mind” is proposed for the secure operation of power grids based on cutting-edge artificial intelligence (AI) technologies. The proposed platform provides a data-driven, model-free and closed-loop control agent trained using deep reinforcement learning (DRL) algorithms by interacting with massive simulations and/or real environment of a power grid. The proposed agent learns from scratch to master the power grid voltage control problem purely from data. It can make autonomous voltage control (AVC) strategies to support grid operators in making effective and timely control actions, according to the current system conditions detected by real-time measurements from supervisory control and data acquisition (SCADA) or phasor measurement units (PMUs). Two state-of-the-art DRL algorithms, namely deep Q-network (DQN) and deep deterministic policy gradient (DDPG), are proposed to formulate the AVC problem with performance compared. Case studies on a realistic 200-bus test system demonstrate the effectiveness and promising performance of the proposed framework.

Subfields: Control and Systems Engineering, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/ed2bcfa4536eee4ed23d26e45e0d2a9af573567b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Control and Systems Engineering

1. Frede Blaabjerg
2. Zhe Chen
3. Yujie Tang
4. Xin Chen
5. Guannan Qu
6. Steven H. Low
7. Na Li
8. Xiongfei Wang

In Electrical and Electronic Engineering

1. Na Li
2. Steven H. Low
3. Yujie Tang
4. Zhe Chen
5. Frede Blaabjerg
6. Xin Chen
7. Guannan Qu
8. Xiongfei Wang

Citations in the corpus

• Reinforcement Learning for Selective Key Applications in Power Systems: Recent Advances and Future Challenges (2021, methodology): “For example, in [90], [91], [95], the reward is defined as (23), An alternative is to set the reward to be negative (e.g., − 1 ) when the voltage is outside an acceptable range (e.g., ± 5% of the nominal value), and positive (e.g., +1 ) when inside the range [94].”
• Reinforcement Learning and Its Applications in Modern Power and Energy Systems: A Review (2020, methodology): “In [100], the DDPG algorithm is also applied to train an agent to act as “grid mind” for the secure operation of power grid.”