Towards Human-Steerable Battery Energy Storage System Optimization: A Novel MDP Framework | AASG Workshop @ AAMAS 2024

Abstract

This paper introduces a novel formulation for optimization of the BESS (battery energy storage system) problem, a crucial compo- nent for driving renewable energy production to profitability. We review the existing literature on several optimization methods and make a case for the need of this novel MDP formulation that will be a foundation for learning from human demonstrations and feed- back, prioritizing different constraints and real-world situations in addition to optimizing for profitability.

Cite

@inproceedings{bessrl2024,
      title={Towards Human-Steerable Battery Energy Storage System Optimization: A Novel MDP Framework}, 
    author={
        Sai Krishna Gottipati and 
        Clod{\'e}ric Mars and 
        Julien Gabaud and
        Vahid Abdollahi and
        Laila El Moujtahid and
        Matthew E. Taylor
    },
    booktitle={5th International Workshop on Autonomous Agents for Social Good},
    year={2024}
}
Previous
Previous

GLIDE-RL: Grounded Language Instruction through DEmonstration in RL | AAMAS 2024

Next
Next

Curriculum Learning for Cooperation in Multi-Agent Reinforcement Learning | ALOE Workshop @ NeurIPS 2023