Green Hydrogen under Uncertainty: Evaluating Power-to-X Strategies Using Agent-Based Simulation and Multi-Criteria Decision Framework

TL;DR

This paper introduces a novel agent-based and multi-criteria decision framework to evaluate green hydrogen production strategies under real-world market uncertainties, emphasizing dynamic actor interactions and system performance.

Contribution

It presents a new integrated modeling approach combining agent-based simulation with multi-criteria decision analysis for Power-to-X strategies, addressing limitations of static models.

Findings

Hybrid strategies outperform in cost and emissions.

On-site-only strategies minimize costs and emissions but can't meet demand.

Dynamic actor interactions significantly influence system performance.

Abstract

The transition toward net-zero energy systems requires scalable and cost-effective deployment of Power-to-X technologies, particularly green hydrogen production. Despite increasing investments, a critical research gap remains in dynamically assessing how different operational strategies affect the feasibility of hydrogen production under real-world energy market conditions. Most existing studies rely on static, techno-economic models and overlook actor interactions, infrastructure limitations, and regulatory complexity. This paper presents a novel modeling framework that integrates agent-based simulation with multi-criteria decision-making to evaluate green hydrogen production strategies using co-located wind and solar generation. Three operational strategies - grid-only, on-site-only, and hybrid - are applied across three electrolyzer capacity levels (10 MW, 50 MW, and 100 MW) within a Danish case study. Real electricity tariffs, emissions factors, and market data are used to simulate technical, economic, and environmental performance indicators. The results show that hybrid strategies consistently outperform grid-only configurations in terms of cost and emissions while maintaining stable hydrogen output. Although on-site-only strategies minimize emissions and costs, they fail to meet fixed production demands. This framework offers novel scientific contributions by modeling dynamic actor interactions and integrating system performance evaluation into strategic planning. Practically, it provides actionable insights for energy planners and policymakers designing resilient and efficient Power-to-X systems in renewable-rich contexts.