Derivation of one- and two-factor experience curves for electrolysis technologies

Green hydrogen is seen as a promising energy carrier contributing to the decarbonization of the energy system. The competitiveness of green hydrogen compared to conventional produced (grey) hydrogen strongly depends on the investment cost of electrolysers and the costs of required green electricity. Accordingly, the expected investment costs and the efficiency of the production process for electrolysis technologies will play a decisive role. Experience curves can help estimate these crucial key parameters better. This paper applies experience curves to electrolysis technologies to enhance the understanding of these developments, specifically for alkaline, proton exchange membranes and solid oxide electrolysis. Experience rates are estimated using one- and two-factor experience curves for CAPEX and one-factor experience curves for electricity consumption. An independent and extensive database of CAPEX, electrical consumption and cumulative installed capacity was developed. Additionally, two different time frames and different data handling techniques are applied to understand the impact on the experience rates and the fit of the experience curve to the data provided. Based on the extensive database developed, CAPEX experience rates for the technologies range from 7%–21%, depending on the technology. In almost all cases, a two-factor experience curve analysis reduces the experience rate based on the cumulative installed capacity with the experience rate based on the size ranging between 7.4-10.2 % for all technologies. Lastly, the experience curve of the electrical consumption expects a reduction of around 1-3.5 % with each doubling of the capacity. • Derivation of experience rates for the major electrolysis technologies (AEL, PEM, SOEC). • Calculation of one- and two-factor experience rates for CAPEX with considerable impact of sizes as a second factor. • Calculation of one-factor experience rates for electricity consumption, with results between 1-3.5 %. • In depth consideration of the effects on the experience rates of data handling techniques and time frame variation. • Estimating learning investment to reach CAPEX targets for electrolysers of the U.S. Department of Energy.