Litcius/Paper detail

Research on Control Strategy of Hybrid Superconducting Energy Storage Based on Reinforcement Learning Algorithm

Yang Liu, Xingfan Han, Zuoxia Xing, Pengtao Li, Hengyu Liu, Zhanpeng Jiang

2024IEEE Transactions on Applied Superconductivity18 citationsDOI

Abstract

Frequent battery charging and discharging cycles significantly deteriorate battery lifespan, subsequently intensifying power fluctuations within the distribution network. This paper introduces a microgrid energy storage model that combines superconducting energy storage and battery energy storage technology, and elaborates on the topology design and energy management strategy of the model in detail. Concurrently, this paper delve into the operational principles and control mechanisms of the hybrid energy storage system. To enhance the performance of microgrid energy storage model, a reinforcement learning algorithm is proposed to design the optimal strategy. In addition, the feasibility of the energy storage model was verified through rigorous simulation analysis. The research results indicate that hybrid energy storage systems promote more stable operation of the power grid, thereby improving the reliability of the power system.

Topics & Concepts

Reinforcement learningSuperconducting magnetic energy storageComputer scienceSuperconductivityEnergy (signal processing)AlgorithmSuperconducting magnetArtificial intelligenceCondensed matter physicsPhysicsQuantum mechanicsFrequency Control in Power SystemsPower Systems and Renewable EnergyMicrogrid Control and Optimization