energy-sdk-l2rpn (https://github.com/NVIDIA/energy-sdk-l2rpn): A Graphical User Interface (GUI) to find some interesting actions, based on simulation
grid2viz (https://github.com/rte-france/grid2viz): A GUI to inspect a posteriori the behaviour of some agents
grid2game (https://github.com/BDonnot/grid2game/ ): A GUI that allows to manually interact with an agent and to "play the game" manually
chronix2grid (https://github.com/BDonnot/ChroniX2Grid ): A software allowing to generate simulated data that can be used by grid2op (in particular)
If you developed a something related to the grid2op ecosystem, please let us know and we'll add it here.
PAPERS TO GO FURTHER - Join us!
Below is the list of accepted and submitted papers from our Apogee project Team since 2017.
It should help you have a deeper understand of the challenges we are tackling and yet the approaches we have been exploring. We hope this could give you draw some inspirations to join us make further advances in the field of AI for Smart Grids!
Most of those papers can be found on arxiv and HAL libraries.
NeurIPS 2022 (RL4RealLife Workshop) : Power Grid Congestion Management via Topology Optimization with AlphaZero.
Challenge's winner in 2022. This paper presents how an DRL algorithm inspired from AlphaZero is used to tackle the congestion management in powergrid.
The challenge's winner in 2020. This paper uses Actor-Critic methods and approximate value functions via GNNs.
PCML- NeurIPS 2020: Learning to run a Power Network Challenge: a Retrospective Analysis
The competition's description and results in 2020.
Simple baseline approach using RL to represent an artificial control room operator that can operate a IEEE 14-bus test case for a duration of 1 week.
A decentralized approach to the problem using multi-agent RL.
An adversarial training approach for injecting robustness.
2019 L2RPN participant. Imitation learning is used to provide a good initial policy. Then, the agent is trained with DRL algorithms to improve its policy and an Early Warning (EW) mechanism is designed to help the agent find good topology control strategies for long testing periods.
Neurocomputing Jounal: LEAP Nets for System Identification and Application to Power Systems
PSCC 2020: Unsupervised Graph Neural Solver for Power Flow Computation
Accepted papers 2017 -2019
ISGT Europe 2018: Optimization of computational budget for power system risk assessment
ISGT Europe 2018 & NIPS 2019 Workshop: Guided machine learning for power grid segmentation
ESANN 2019: Leap Net for power grid perturbations
IJCNN 2019: Graph Neural Solvers for Power Systems
EGC 2019: Semi-supervised labelling, Towards an Extended Expert Approch
Réseau de Transport d'Électricité (Electricity Transmission Network), usually known as RTE, is the electricity transmission system operator of France. It is responsible for the operation, maintenance and development of the French high-voltage transmission system, which at approximately 100,000 kilometres (62,000 mi), is Europe's largest. RTE R&D is one of the strongest in the world in the field of power grids and has many research collaboration around the world, especially in Europe and in the USA. RTE is now a member of the Linux Foundation Energy Initiative in which it open-sourced many simulators and applications.
ChaLearn is a non-profit organization with vast experience in the organization of academic challenges. ChaLearn is interested in all aspects of challenge organization, including data gathering procedures, evaluation protocols, novel challenge scenarios (e.g., competitions), training for challenge organizers, challenge analytics, result dissemination and, ultimately, advancing the state-of-the-art through challenges.