Representation Learning Enhanced Deep Reinforcement Learning for Optimal Operation of Hydrogen-based Multi-Energy Systems

TL;DR

This work tackles the challenge of optimally operating hydrogen-based multi-energy systems (HMES) with nonlinear, multi-physics hydrogen energy storage dynamics and multiple uncertainties. It introduces a representation-learning enhanced DRL (SR-DRL) framework that incorporates state and state-action embeddings to restructure the state space and accelerate policy learning within an actor-critic setup (DDPG/TD3). Through real-world data experiments, the approach demonstrates faster convergence, higher-quality policies, reduced operating costs, and improved constraint satisfaction compared to conventional DRL, with safety-critical HESS variables explicitly modeled. The results suggest SR-DRL as a practical, scalable method for intelligent, safe HMES control, with potential for extension to multi-agent coordination in interconnected energy networks.

Abstract

Hydrogen-based multi-energy systems (HMES) have emerged as a promising low-carbon and energy-efficient solution, as it can enable the coordinated operation of electricity, heating and cooling supply and demand to enhance operational flexibility, improve overall energy efficiency, and increase the share of renewable integration. However, the optimal operation of HMES remains challenging due to the nonlinear and multi-physics coupled dynamics of hydrogen energy storage systems (HESS) (consisting of electrolyters, fuel cells and hydrogen tanks) as well as the presence of multiple uncertainties from supply and demand. To address these challenges, this paper develops a comprehensive operational model for HMES that fully captures the nonlinear dynamics and multi-physics process of HESS. Moreover, we propose an enhanced deep reinforcement learning (DRL) framework by integrating the emerging representation learning techniques, enabling substantially accelerated and improved policy optimization for spatially and temporally coupled complex networked systems, which is not provided by conventional DRL. Experimental studies based on real-world datasets show that the comprehensive model is crucial to ensure the safe and reliable of HESS. In addition, the proposed SR-DRL approaches demonstrate superior convergence rate and performance over conventional DRL counterparts in terms of reducing the operation cost of HMES and handling the system operating constraints. Finally, we provide some insights into the role of representation learning in DRL, speculating that it can reorganize the original state space into a well-structured and cluster-aware geometric representation, thereby smoothing and facilitating the learning process of DRL.

Figures (7)

• Figure 1: Hydrogen-based multi-energy system for a building community
• Figure 2: The architecture of representation learning enhanced DRL method
• Figure 3: The evolution of episode return during training under different methods.
• Figure 4: Distribution of stacked cost and penalty of constraint violations under different methods.
• Figure 5: The evolution of SOC of different types of energy storage devices in response to electricity market price with different methods.