Logo image
A Double Deep Q-Learning Based Real-Time Energy Management Optimization for Microgrids
Conference proceeding

A Double Deep Q-Learning Based Real-Time Energy Management Optimization for Microgrids

Xiangru Shi, Flavie Didier, Abderrezak Badji, Sheikh Izzal Azid, Maurizio Cirrincione and Salah Laghrouche
2025 IEEE International Conference on Energy Technologies for Future Grids (ETFG), pp.1-6
2025 IEEE International Conference on Energy Technologies for Future Grids (ETFG) (Wollongong, Australia, 07/12/2025–11/12/2025)
2025

Abstract

Batteries Costs Double deep Q network Energy Management System (EMS) Energy management systems grid-connected microgrid Microgrids Optimization Real-time optimization Real-time systems Testing Thermal stability Training Uncertainty
The uncertainties introduced by distributed energy resources bring significant challenges to the real-time operation of microgrids, particularly in managing the fluctuating generation and load profiles required for optimal energy management system (EMS) performance. This study proposes a deep reinforcement learning-based framework to achieve real-time scheduling of battery storage within a grid-connected microgrid. The control problem is formulated as a Markov Decision Process (MDP), a Double Deep Q-Network (DDQN) agent is trained to generate optimal charging and discharging actions based on observed system states. Simulation results demonstrate that the proposed EMS controller can effectively derive optimal control policies, ensuring system stability, cost efficiency, and energy selfsufficiency.

Details

Metrics

1 Record Views
Logo image