Learning to run a power network challenge for training topology controllers

Marot, Antoine; Donnot, Benjamin; Romero, Camilo; Donon, Balthazar; Lerousseau, Marvin; Veyrin-Forrer, Luca; Guyon, Isabelle

doi:10.1016/j.epsr.2020.106635

Cited by 46 publications

(79 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dataset for the IEEE 14-bus model contains 1000 scenarios with data for 28 continuous days in a 5minute interval. Each scenario includes pre-determined load variations and generation schedules which are representative of distributions of the French grid [7]. In this study only the first week of each scenario will be used.…”

Section: Power System Frameworkmentioning

confidence: 99%

“…We note that the formulation of the control problem considered here as a Markov Decision Process (MDP) can be found e.g. in [7], [12].…”

Section: Reinforcement Learning Approachmentioning

confidence: 99%

“…Finally, we use the same reward function as defined in [7], [12]. That is, the immediate reward at each time step reflects the remaining available transfer capabilities.…”

Section: B Observations Neural Network Rewardmentioning

confidence: 99%

“…Note that this indirectly favours successful operation of the grid, because (i) overloads are discouraged and (ii) a "game over" case stops the accumulation of reward. For more details we refer to [7], [12].…”

Section: B Observations Neural Network Rewardmentioning

confidence: 99%

“…This competition was conducted with the primary goal of introducing and recognising the potential of AI and ML based tools to support control rooms and assist in making optimal decisions. The first edition of this competition was held in 2019 [7]. The L2RPN competition focuses on learning a subset of actions that can be taken by grid operators, namely topological changes.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Exploring grid topology reconfiguration using a simple deep reinforcement learning approach

Subramanian

Viebahn

Tindemans

et al. 2021

2021 IEEE Madrid PowerTech

Self Cite

View full text Add to dashboard Cite

System operators are faced with increasingly volatile operating conditions. In order to manage system reliability in a cost-effective manner, control room operators are turning to computerised decision support tools based on AI and machine learning. Specifically, Reinforcement Learning (RL) is a promising technique to train agents that suggest grid control actions to operators. In this paper, a simple baseline approach is presented using RL to represent an artificial control room operator that can operate a IEEE 14-bus test case for a duration of 1 week. This agent takes topological switching actions to control power flows on the grid, and is trained on only a single well-chosen scenario. The behaviour of this agent is tested on different timeseries of generation and demand, demonstrating its ability to operate the grid successfully in 965 out of 1000 scenarios. The type and variability of topologies suggested by the agent are analysed across the test scenarios, demonstrating efficient and diverse agent behaviour.

show abstract

Section: Power System Frameworkmentioning

confidence: 99%

“…We note that the formulation of the control problem considered here as a Markov Decision Process (MDP) can be found e.g. in [7], [12].…”

Section: Reinforcement Learning Approachmentioning

confidence: 99%

“…Finally, we use the same reward function as defined in [7], [12]. That is, the immediate reward at each time step reflects the remaining available transfer capabilities.…”

Section: B Observations Neural Network Rewardmentioning

confidence: 99%

Section: B Observations Neural Network Rewardmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations