Source-linked AI summary
On the economics of electrical storage for variable renewable energy sources
Alexander Zerrahn, Wolf-Peter Schill, Claudia Kemfert
TL;DR
The paper examines whether high storage requirements could constrain the expansion of variable wind and solar power. It relaxes corner-solution assumptions and uses a parsimonious optimization model with an economic objective. Storage needs are up to two orders of magnitude lower, leading the paper to conclude that storage is unlikely to limit the renewable transition.
Problem
Sinn’s analysis suggests that excessive electrical storage requirements may impede further expansion of variable wind and solar power.
Method
The paper combines storage with renewable curtailment and uses a parsimonious optimization model to determine economically efficient capacity and curtailment choices.
Results
Storage needs are lower by up to two orders of magnitude when the strong assumptions in Sinn’s approach are addressed.
Takeaways & Limitations
Electrical storage is unlikely to limit the transition to renewable energy.
Takeaways & Limitations
The analysis abstracts from many flexibility options and its power-to-x effects depend strongly on assumptions about capacity and full-load hours.
Abstract
from arXiv · showhide
The use of renewable energy sources is a major strategy to mitigate climate change. Yet Sinn (2017) argues that excessive electrical storage requirements limit the further expansion of variable wind and solar energy. We question, and alter, strong implicit assumptions of Sinn's approach and find that storage needs are considerably lower, up to two orders of magnitude. First, we move away from corner solutions by allowing for combinations of storage and renewable curtailment. Second, we specify a parsimonious optimization model that explicitly considers an economic efficiency perspective. We conclude that electrical storage is unlikely to limit the transition to renewable energy.
1. Introduction
The paper argues that variable renewable energy creates supply–demand timing challenges, but Sinn’s extreme storage assumptions overstate the requirements. Allowing curtailment and optimizing economic trade-offs yields substantially lower storage needs and supports continued renewable expansion.
- Wind and solar output varies with weather, time, season, and location, creating mismatches between renewable supply and electricity demand.
- Sinn’s corner cases either store all surplus renewable energy or store none, producing either excessive storage needs or substantial unused generation.
- An economically efficient solution combines storage with some renewable curtailment rather than choosing between the two extremes.
- The paper uses a parsimonious optimization model with an economic objective that trades off storage, renewable curtailment, and other power-market assets.
- Storage needs are up to two orders of magnitude lower when moving away from corner solutions, and moderate in the economically optimized model.
- The paper concludes that electrical storage is unlikely to limit the transition to renewable energy.
2. Literature review
The literature review finds that Sinn’s storage estimates are unusually high relative to established studies. It links the difference to renewable curtailment and broader flexibility trade-offs that reduce the need for storage.
- Storage needs grow disproportionately with higher renewable shares because surplus energy is concentrated in a few high-peak hours.
- Across reviewed studies, storage requirements are often at least an order of magnitude lower than Sinn’s estimates.
- The review identifies renewable curtailment as a key reason literature estimates differ from Sinn’s corner-solution results.
- Combining renewable capacity oversizing with temporary curtailment can substitute for storage expansion under economic-efficiency criteria.
- Other flexibility options include geographical balancing, demand-side management, and flexible renewable use in heat or mobility sectors.
- Sinn’s findings are outliers compared with the established literature on storage requirements for high shares of variable renewables.
3. Replication and intuition
The authors replicate Sinn’s central storage findings with open data and software, then examine why storage requirements vary and how the model’s assumptions affect practical relevance.
- Model assumptions: Sinn’s approach seeks to integrate all renewable generation through storage, but lacks explicit economic or optimality criteria.
- Replication: Open data and an open-source software tool reproduce Sinn’s central findings for variable-renewable shares between 20% and 90%.
- Replication: At 50% variable renewables, required storage reaches 2.1 TWh, two orders of magnitude above current German pumped-hydro capacity.
- Robustness: Storage requirements are highly sensitive to the input-data base year; 2014 produces the highest needs up to 65% renewables, while 2015 produces the smallest.
- Intuition: Residual load equals hourly demand minus renewable feed-in, and its duration curve reveals both high-demand deficits and renewable-surplus hours that affect storage needs.
4. Storage requirements under renewable curtailment
Allowing renewable curtailment creates intermediate solutions between no curtailment and no storage. Curtailing surplus reduces storage requirements substantially, with moderate increases in renewable capacity and no increase in backup-capacity needs.
- Power-oriented curtailment: Increasing renewable curtailment lowers storage requirements, with the decline close to linear but somewhat convex.
- Power-oriented curtailment: At 50% variable renewables, 5% curtailment reduces required storage from 2.1 TWh to 0.3 TWh.
- Combined solutions: The model combines storage and curtailment between the corner cases of no curtailment and no storage.
- Capacity trade-off: Achieving 50% renewable energy requires 214 GW without curtailment versus 226 GW with 5% curtailment.
- Backup capacity: Curtailment does not increase the backup capacities needed to supply residual demand after storage.
- Energy-oriented curtailment: Energy-oriented curtailment targets hours that produce the greatest reductions in storage energy capacity.
5. Cost-minimal storage requirements
The cost-minimization model allows renewable curtailment, storage, and conventional generation to coexist, rather than restricting solutions to corner cases. It finds moderate storage needs, with storage energy capacity and curtailment increasing as renewable shares rise, while flexible power-to-x demand can reduce both.
- Model and economic perspective: The model minimizes cost across conventional and renewable generation, renewable curtailment, and electrical storage, including storage energy and power capacities.It represents an economic first-best perspective under perfect competition and complete information, with stylized conventional and variable-renewable technologies.
- Model and economic perspective: Allowing curtailment produces interior solutions that combine storage, renewables, conventional plants, and sometimes curtailed renewable generation.This replaces the corner solutions of no renewable curtailment or no storage considered in Sinn’s approach.
- Storage requirements: The energy-to-power ratio rises from about 6 hours at 50% renewables to about 19 hours at 90%, while no storage for weeks or months is needed.Optimal storage energy and power capacities both increase with the renewable share, but energy capacity grows faster.
- Curtailment: Optimal renewable curtailment increases with the minimum renewable share, because avoiding all curtailment is not always cost-efficient.The cost-optimal system combines conventional plants, storage, and renewables, with some renewable output curtailed at times.
- Flexible sector coupling: With flexible power-to-x demand, storage energy capacity at 50% renewables falls from 35 GWh to 4 GWh and curtailment falls from 5% to below 1%.Storage requirements remain lower with power-to-x technologies up to an 85% renewable share; flexible demand can absorb renewable surplus when availability is high.
6. Discussion
The discussion identifies several reasons storage needs may be lower than estimates based on stylized assumptions, including cost-efficient mixes, differentiated technologies, broader markets, and flexible demand.
- Storage needs depend on modeling assumptions, including power-to-x parameters, base-year data, and treatment of external costs and regulation.The discussion notes that power-to-x effects depend strongly on capacities and full-load hours, while external-cost shares and regulatory requirements remain for further analysis.
- Cost-efficient solutions combine renewable expansion, curtailment, and electrical storage rather than relying on storage alone.
- A differentiated storage fleet can address intra-hourly, diurnal, and seasonal fluctuations while tending to be smaller and cheaper.
- Historical wind and solar profiles can overestimate flexibility requirements because technology choices and generation patterns affect variability.
- Broader electricity-market integration smooths residual load and lowers storage requirements.
- More temporally flexible demand can substitute for electrical storage and enable consumers to capture arbitrage gains.
7. Conclusions
The conclusions revisit claims that storage could constrain renewable expansion and report substantially lower needs when the analysis allows economically efficient combinations of renewables, curtailment, and storage.
- Storage needs are lower by up to two orders of magnitude than in the compared framework.
- Cost-efficient solutions optimally combine renewable capacity expansion, renewable curtailment, and electrical storage.
- The analysis allows flexible additional demand from heating, mobility, or hydrogen production, which may further decrease electrical storage needs.The paper demonstrates that power-to-x options may substantially change the picture but calls for more detailed research.
- The paper concludes that electrical storage requirements do not limit the further expansion of variable renewable energy sources.
Appendix A. Literature review: storage power capacity requirements
The literature review compares storage power-capacity requirements across studies and reports that the literature generally finds much lower capacities than Sinn (2017).
- The literature finds much lower storage power capacities than Sinn (2017).
- The comparison assembles storage-power estimates from multiple studies and selects relevant cases when underlying information is incomplete.Missing data are calculated or inferred for some studies, including peak load or yearly demand.
- The reviewed cases span different technologies, scenarios, renewable shares, and storage assumptions.
Appendix B. Sensitivity: optimal storage for different base years
The sensitivity analysis shows that cost-minimal storage capacities vary substantially with the selected base year.
- Cost-minimal storage capacities are highly sensitive to the choice of base year.
- Base years 2015 and 2016 deliver much lower optimal storage capacities than the other base years.Smoother residual-load patterns are identified as an important factor driving this result.
- Cost-minimal capacities differ substantially from calculations that do not take costs into account.
Appendix C. Storage requirements in the optimization model: sensitiv-
The lithium-ion sensitivity preserves the model’s qualitative conclusions while changing the optimal storage mix and capacities. Compared with pumped hydro, batteries require less storage energy but entail higher curtailment and slightly higher electricity costs.
- Qualitative results are unchanged, and storage energy requirements remain substantially lower than in Sinn (2017).
- Lithium-ion storage has lower storage-energy deployment than pumped hydro because its specific costs are higher.
- Curtailment is higher with lithium-ion batteries: 6% versus 5% at 50% renewables and 21% versus 16% at 80%.
- Electricity-provision costs are slightly higher when pumped-hydro storage is unavailable, while pumped hydro appears appropriate for this stylized analysis.
- A cost-minimal real-world portfolio would combine storage technologies with other flexibility options, but the paper does not develop a full analysis.
Appendix D. Power-to-x: sensitivity with respect to different configura-
The sensitivity analysis examines how flexible power-to-x demand affects optimal storage requirements. Medium full-load-hour configurations can reduce storage needs, while very high full-load hours increase them again because demand and renewable availability become more mismatched.
- Full-load hours between around 1,000 and 3,500 can trigger substantially lower storage needs.
- At lower full-load hours, power-to-x demand uses renewable surpluses that would otherwise be curtailed, producing little or no change in optimal storage capacity.
- At very high full-load hours, storage needs increase again as power-to-x demand becomes more mismatched with renewable availability.
- The mismatch drives disproportionate renewable-capacity expansion, which increases curtailment and the optimal amount of electrical storage.
- Figure D.1 reports storage requirements across power-to-x settings for 70% variable renewables, with the largest impact between 1,500 and 2,000 full-load hours.