Resilient energy management in hydrogen-integrated microgrids using actor-critic reinforcement learning

Tai Zhang

doi:10.1016/j.apenergy.2026.128568

Resilient energy management in hydrogen-integrated microgrids using actor-critic reinforcement learning

2026-08-10

0

OA

AI

T

Tai Zhang

S

Spyros Giannelos

D

Danny Pudjianto

G

Goran Štrbac *

DOI：10.1016/j.apenergy.2026.128568

Original

Original request for help

Save

Abstract

En 中文

• Physics-embedded AC-PPO supports feasibility-aware hydrogen-microgrid control. • AC-PPO avoids blackout events in the severe in-distribution benchmark. • Genuine out-of-distribution tests show graceful performance degradation. • Public weather and building-load data support robustness without retraining. • Reward-weight sensitivity analysis identifies a stable safety-oriented region.

Keywords:

Reinforcement learning

Energy resilience

Hydrogen energy storage

Microgrid management

Uncertainty modelling

Resilient energy management in hydrogen-integrated microgrids using actor-critic reinforcement learning

Abstract

Journal

Organization

Cited Papers

Citing Papers