ePrints

Energy management for renewable power-to-ammonia in networked multi-energy microgrids: A multi-agent Soft Actor–Critic approach

Lookup NU author(s): Dr Sheng WangORCiD

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.

Abstract

© 2026 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.Renewable power-to-ammonia (RePtA) has gained increasing interest in networked multi-energy microgrids (MEMGs) due to its potential as a zero-carbon fuel. This paper presents an energy management model for networked MEMGs with RePtA considering laddered carbon trading mechanism. First, we establish a RePtA system model integrating renewable hydrogen/ammonia production, storage, and utilization, with thermo-electro-chemical synthesis dynamics. Second, we frame the renewable energy fluctuation challenge as a Markov Decision Process under unknown transition dynamics. Third, we introduce a novel model-free Multi-Agent Deep Reinforcement Learning (MADRL) algorithm featuring a centralized training and decentralized execution (CTDE) architecture. This framework leverages the Soft Actor–Critic (SAC) approach to derive real-time optimal control policies under stochastic renewable generation. Case studies validate the superiority of the proposed method against other state-of-the-art MADRL algorithms.

Publication metadata

Author(s): Song D, Yan L, Zhu D, Wang S, Li Z, Pouresmaeil E, Zhai J

Publication type: Article

Publication status: Published

Journal: International Journal of Hydrogen Energy

Year: 2026

Volume: 245

Print publication date: 24/06/2026

Online publication date: 29/05/2026

Acceptance date: 24/05/2026

ISSN (print): 0360-3199

ISSN (electronic): 1879-3487

Publisher: Elsevier Ltd

URL: https://doi.org/10.1016/j.ijhydene.2026.155742

DOI: 10.1016/j.ijhydene.2026.155742

Altmetrics

Share