Coordinated Dispatch of Energy Storage Systems in the Active Distribution Network: A Complementary Reinforcement Learning and Optimization Approach

TL;DR

This paper introduces a hybrid reinforcement learning and optimization method for real-time, safe, and economical dispatch of energy storage systems in active distribution networks, addressing nonlinearity and renewable energy variability.

Contribution

It presents a novel SA2CO approach combining RL and optimization with a neural network for rapid security assessment, improving dispatch performance in active distribution networks.

Findings

Effective in real-time dispatch of multiple ESSs

Outperforms existing RL and optimization methods

Demonstrates scalability and security in simulations

Abstract

The complexity and nonlinearity of active distribution network (ADN), coupled with the fast-changing renewable energy (RE), necessitate advanced real-time and safe dispatch approach. This paper proposes a complementary reinforcement learning (RL) and optimization approach, namely SA2CO, to address the coordinated dispatch of the energy storage systems (ESSs) in the ADN. The proposed approach leverages RL's capability to make fast decision and address the model inaccuracies, while optimization methods ensure the ADN security. Furthermore, a hybrid data-driven and expert-experience auxiliary neural network is formulated as a rapid security assessment component in the SA2CO algorithm, enabling dynamic switching between RL and optimization methodologies. Simulation results demonstrate the proposed method's effectiveness and scalability in achieving real-time, safe, and economical dispatch of multiple ESSs in the ADN, surpassing the performance of the state-of-the-art RL and optimization methods.