Curriculum Based Reinforcement Learning of Grid Topology Controllers to Prevent Thermal Cascading

TL;DR

This paper presents a curriculum-based reinforcement learning approach with domain knowledge integration to control grid topology and prevent thermal cascading, achieving high accuracy and speed in real-world power system scenarios.

Contribution

It introduces a novel actor-critic RL agent with curriculum learning and parallel scenario training, tailored for power grid topology control to prevent cascading failures.

Findings

Achieved 2nd place in accuracy and 1st in speed in a real-world challenge.

Demonstrated the importance of domain knowledge integration for effective RL in power systems.

Developed open-source code for practical application and further research.

Abstract

This paper describes how domain knowledge of power system operators can be integrated into reinforcement learning (RL) frameworks to effectively learn agents that control the grid's topology to prevent thermal cascading. Typical RL-based topology controllers fail to perform well due to the large search/optimization space. Here, we propose an actor-critic-based agent to address the problem's combinatorial nature and train the agent using the RL environment developed by RTE, the French TSO. To address the challenge of the large optimization space, a curriculum-based approach with reward tuning is incorporated into the training procedure by modifying the environment using network physics for enhanced agent learning. Further, a parallel training approach on multiple scenarios is employed to avoid biasing the agent to a few scenarios and make it robust to the natural variability in grid operations. Without these modifications to the training procedure, the RL agent failed for most test scenarios, illustrating the importance of properly integrating domain knowledge of physical systems for real-world RL learning. The agent was tested by RTE for the 2019 learning to run the power network challenge and was awarded the 2nd place in accuracy and 1st place in speed. The developed code is open-sourced for public use.