Mr Oliver Schmidt, Imperial College London, United Kingdom
Dr Iain Staffell, Imperial College London,United Kingdom
Dr Sheridan Few, Imperial College London, United Kingdom
Mr Ajay Gambhir, Imperial College London, United Kingdom
High uncertainty regarding the potential future costs of energy storage technologies negatively impacts energy systems modelling and economic analyses of low-carbon energy systems .
An understanding of the sources of innovation for the most promising energy storage technologies and their potential future costs is therefore highly desirable and would enrich the sessions of this year’s BIEE conference on innovation and opportunities in the energy sector.
We form this understanding by combining three methods for a selection of promising storage technologies:
A. Expert elicitations: We perform a semi-structured interviews with experts from academia and industry on innovation potential and future costs of their technology of expertise.
B. Learning curves: We compile reported learning rates and derive new ones where possible to analyse the required additional installed capacity for cost-competitiveness.
C. Bottom-up models: We model the impact of possible innovations, based on the expert elicitations and the literature, on the bill of materials and manufacturing costs.
Preliminary results of the expert elicitations on batteries and electrolysis systems indicate that future cost reductions are less likely a result of further innovation in R&D, but of improvements in manufacturing due to scale-up. The results show which innovations in R&D and manufacturing scale-up are deemed feasible by 2020 and 2030 and the range of cost estimates made by the experts for 2020 and 2030, based on these possible innovations.
The analysis of the learning curves allows us to identify a current cost benchmark. The corresponding technologies exhibit learning rates between 2% and 4% [2, 3]. Alternative battery technologies show learning rates between 9% and 22% at slightly higher costs, but lower cumulative installed capacity [4, 5, 6]. Electrolysis and fuel cells show learning rates of 18% and 15% respectively . For technologies with too few data points to derive learning rates, we show current costs and cumulative installed capacity. The obtained learning curves allow us to calculate the required additional capacity to reach the current cost benchmark and compare this on a time-scale using growth projections.
By comparing the results of all three methods, we derive likely ranges of future costs for the most promising energy storage technologies.
These findings give us a solid footing from which to inform economic analyses of low-carbon energy systems and serve as input parameters for energy systems modelling. Given the high uncertainty on future cost reductions, cost assumptions based on our analysis are likely to shed more light on the possible role for storage in future energy systems. As next step, we use these future cost assumptions to explore the possible role of storage in our energy systems model.
References: http://energysuperstore.org/esrn/latest-news/e-highways2050-predicting-europes-storage-needs/  Jamasb, T., 2007. The Energy Journal, 28(3), pp.51–71.  Matteson, S. & Williams, E., 2015. Energy Policy, 85(0), pp.71–79.  Nykvist, B. & Nilsson, M., 2015. Nature Climate Change, 5(4), pp.329–332.  Gerssen-Gondelach, S.J. & Faaij, A.P.C., 2012. Journal of Power Sources, 212, pp.111–129.  Matteson, S. & Williams, E., 2015. Technological Forecasting and Social Change, 92, pp.322–331.  Schoots, K. et al., 2008. International Journal of Hydrogen Energy, 33(11), pp.2630–2645.
Key Word Set:
Energy Storage, Expert Elicitations, Learning Curves, Bottom-Up Modelling, Cost Projections
Schmidt_StudentPitch_Cost_Projections_for_Electrical_Energy_Storage.pdf 778.57 KB