Endogenous learning for green hydrogen in a sector-coupled energy model for Europe

Code and data of the paper "Endogenous learning of green hydrogen in a sector-coupled energy model for Europe"

Abstract:

To reach climate neutrality by 2050 the European energy system would have to be radically transformed. This raises
the question of how much this transition is going to cost and which technologies should be supported to enable the
transformation. Although many energy system models deal with cost-optimal pathways, they typically neglect either
the dependency of investment costs on the cumulative deployed capacity of the technology, the temporal variability
of renewable energy sources or they only depict individual sectors. In this study, we want to address these weak
points. We use a European energy system model with full sector-coupling, renewable variability and with endogenous
temporal-delayed learning on renewable generation technologies (solar PV, onshore and offshore wind) and hydrogen
electrolysis. Three different CO2 budgets corresponding to +1.5o C, +1.7o C, +2.0o C temperature increase are applied.
First we show that for a +1.5o C budget scenario it is optimal to deploy hydrogen electrolysis at large scale already
by 2030, which reduces the electrolysis investment cost to 75 e/kWelec by 2050. Second, we compare three different
methods to represent cost reductions and show that modeling without endogenous learning leads to an overestimation of
the marginal prices of hydrogen, especially in the case of ambitious climate targets, of up to 67% in 2030 and 38% in 2050.