Efficient learning of power grid voltage control strategies via model-based deep reinforcement learning (Journal Article) | OSTI.GOV
Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Efficient learning of power grid voltage control strategies via model-based deep reinforcement learning

Journal Article · · Machine Learning

Here this article proposes a model-based deep reinforcement learning (DRL) method to design emergency control strategies for short-term voltage stability problems in power systems. Recent advances show promising results for model-free DRL-based methods in power systems control problems. But in power systems applications, these model-free methods have certain issues related to training time (clock time) and sample efficiency; both are critical for making state-of-the-art DRL algorithms practically applicable. DRL-agent learns an optimal policy via a trial-and-error method while interacting with the real-world environment. It is also desirable to minimize the direct interaction of the DRL agent with the real-world power grid due to its safety-critical nature. Additionally, the state-of-the-art DRL-based policies are mostly trained using a physics-based grid simulator where dynamic simulation is computationally intensive, lowering the training efficiency. We propose a novel model-based DRL framework where a deep neural network (DNN)-based dynamic surrogate model (SM), instead of a real-world power grid or physics-based simulation, is utilized within the policy learning framework, making the process faster and more sample efficient. However, having stable training in model-based DRL is challenging because of the complex system dynamics of large-scale power systems. We addressed these issues by incorporating imitation learning to have a warm start in policy learning, reward-shaping, and multi-step loss in surrogate model training. Finally, we achieved 97.5% reduction in samples and 87.7% reduction in training time for an application to the IEEE 300-bus test system.

Research Organization:
Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
Sponsoring Organization:
USDOE Advanced Research Projects Agency - Energy (ARPA-E)
Grant/Contract Number:
AC05-76RL01830
OSTI ID:
2345752
Report Number(s):
PNNL-SA-172082
Journal Information:
Machine Learning, Vol. 113, Issue 5; ISSN 0885-6125
Publisher:
Springer NatureCopyright Statement
Country of Publication:
United States
Language:
English

References (42)

A comparison of direct and model-based reinforcement learning conference April 1997
A Deep Reinforcement Learning-Based Multi-Agent Framework to Enhance Power System Resilience Using Shunt Resources journal November 2021
A Multi-Agent Deep Reinforcement Learning Method for Cooperative Load Frequency Control of a Multi-Area Power System journal November 2020
Learning and Fast Adaptation for Grid Emergency Control via Deep Meta Reinforcement Learning journal November 2022
Adaptive Power System Emergency Control Using Deep Reinforcement Learning journal March 2020
Imitation Learning journal March 2018
Data-Driven Load Frequency Control for Stochastic Power Systems: A Deep Reinforcement Learning Method With Continuous Action Search journal March 2019
Reactive Power Optimization for Transient Voltage Stability in Energy Internet via Deep Reinforcement Learning Approach journal April 2019
Neural Network Dynamics for Model-Based Deep Reinforcement Learning with Model-Free Fine-Tuning conference May 2018
A power-flow emulator approach for resilience assessment of repairable power grids subject to weather-induced failures and data deficiency journal January 2018
Overcoming Exploration in Reinforcement Learning with Demonstrations conference May 2018
Definition and Classification of Power System Stability – Revisited & Extended journal July 2021
Hybrid approach based on particle swarm optimization for electricity markets participation journal January 2019
Model-augmented safe reinforcement learning for Volt-VAR control in power distribution networks journal May 2022
Power System Emergency Control to Improve Short-Term Voltage Stability Using Deep Reinforcement Learning Algorithm conference September 2019
A generic modeling and development approach for WECC composite load model journal July 2019
Real-time power system generator tripping control based on deep reinforcement learning journal October 2022
Graph Convolutional Network-Based Topology Embedded Deep Reinforcement Learning for Voltage Stability Control journal September 2021
Online Scheduling of a Residential Microgrid via Monte-Carlo Tree Search and a Learned Model journal March 2021
An integrated critic-actor neural network for reinforcement learning with application of DERs control in grid frequency regulation journal October 2019
Reinforcement Learning and Its Applications in Modern Power and Energy Systems: A Review journal January 2020
Surrogate model enabled deep reinforcement learning for hybrid energy community operation journal May 2021
Load Shedding Scheme with Deep Reinforcement Learning to Improve Short-term Voltage Stability conference May 2018
Learning from demonstration and adaptation of biped locomotion journal June 2004
Concepts of undervoltage load shedding for voltage stability journal April 1992
Applications of reinforcement learning in energy systems journal March 2021
Reinforcement learning control approach for autonomous microgrids journal August 2019
Accelerated Derivative-Free Deep Reinforcement Learning for Large-Scale Grid Emergency Voltage Control journal January 2022
Reinforcement Learning for Selective Key Applications in Power Systems: Recent Advances and Future Challenges journal July 2022
An Energy Sector Roadmap to Carbon Neutrality in China book October 2021
Deep-Reinforcement-Learning-Based Autonomous Voltage Control for Power Grid Operations journal January 2020
(Deep) Reinforcement learning for electric power system control and related problems: A short review and perspectives journal January 2019
Model-Free Emergency Frequency Control Based on Reinforcement Learning journal April 2021
Faster than real-time dynamic simulation for large-size power system with detailed dynamic models using high-performance computing platform conference July 2017
Recent and prospective developments in power system control centers: Adapting the digital twin technology for application in power system control centers conference June 2018
Short-Term Voltage Instability: Effects on Synchronous and Induction Machines journal May 2006
Distributional Deep Reinforcement Learning-Based Emergency Frequency Control journal July 2022
Model-free voltage control of active distribution system with PVs using surrogate model-based deep reinforcement learning journal January 2022
Analytic Deep Learning-Based Surrogate Model for Operational Planning With Dynamic TTC Constraints journal July 2021
A post-contingency power flow emulator for generalized probabilistic risks assessment of power grids journal May 2020
Supervised assisted deep reinforcement learning for emergency voltage control of power systems journal February 2022
Deep Belief Network Enabled Surrogate Modeling for Fast Preventive Control of Power System Transient Stability journal January 2022