Techno-Economic and Machine Learning Forecasting of Green Hydrogen: Policy Insights from Costa Rica and the United Kingdom

Green hydrogen is increasingly viewed as a cornerstone of global decarbonization, yet its economic feasibility varies widely due to differences in renewable resource quality, electricity pricing, infrastructure scale, and policy frameworks. This study develops a cross-national modeling framework integrating high-resolution spatial analysis, deterministic techno-economic projections, Monte Carlo simulation, and supervised machine learning to evaluate green hydrogen production in Costa Rica and the United Kingdom. By linking financial modeling, uncertainty quantification, and interpretable learning, the framework estimates Levelized Cost of Hydrogen (LCOH) and Net Present Value (NPV) under region-specific scenarios from 2030 to 2060. Results show that in Costa Rica, Solid Oxide Electrolysis Cell (SOEC) systems at a 3 MW scale achieve a median LCOH of $4.06/kg H₂ by 2030—outperforming PEM and Alkaline systems—and decline to $2.84/kg by 2060. Corresponding NPV values rise from $5.69 million to over $9 million. In the United Kingdom, 50 MW SOEC projects achieve a median LCOH of $3.39/kg in 2030 and $1.39/kg in 2060, with NPVs exceeding $2.1 billion. SHAP-based analysis confirms electricity price, CAPEX, and capacity factor as dominant cost drivers. This simulation-augmented framework provides a transparent tool for hydrogen investment planning under uncertainty. It demonstrates how emerging economies like Costa Rica can pursue decentralized, sustainability-driven hydrogen strategies by aligning with international partners. In particular, the United Kingdom’s offshore wind expertise and financing mechanisms can support Costa Rica’s transition through technical cooperation and policy transfer. The results highlight how cross-national collaboration can accelerate equitable, resilient hydrogen development.