Dynamic optimisation of alkaline electrolyser operation under virtual solar and hybrid solar-wind PPAs – A multi-state MINLP framework applied to the TRIERES hydrogen valley project

This study presents an integrated optimisation framework for the dynamic operation of pressurised alkaline electrolysers powered by hybrid solar-wind Power Purchase Agreements (PPAs). A multi-state, module-level electrolyser model is developed, capturing load-dependent efficiency and incorporating Mixed Integer Non Linear Programming (MINLP) constraints to balance accuracy and computational efficiency. The framework minimises operating costs through optimal dispatch and is applied to a 30 MW electrolyser under the TRIERES hydrogen valley in Greece. Results highlight hydrogen demand as the key driver of flexibility, cost, and emissions. Under low demand, high flexibility yields a Levelised Cost of Hydrogen (LCOH) of 4.79 €/kgH 2 and carbon intensity of 6.33 kgCO 2 /kgH 2 . High demand reduces flexibility, increasing LCOH by 38% and emissions by 70%. Enforcing Renewable Fuels of Non-Biological Origin (RFNBO) compliance raises costs by 7% and LCOH by 14%. Hybrid solar–wind PPAs significantly improve performance, reducing carbon intensity to 3.37 kgCO 2 /kgH 2 and LCOH to 3.66 €/kgH 2 . • Dynamic multi-state electrolyser model with load-dependent efficiency. • MINLP framework for optimal dispatch under hybrid renewable PPAs. • Hydrogen demand drives flexibility, costs, and carbon intensity. • Environmental constraints increase LCOH (+0.65 €/kgH 2 ) but enable RFNBO compliance. • Hybrid solar–wind PPAs lower both LCOH and hydrogen carbon intensity.