Lithium-ion BESS excel at short-term storage needs, addressing daily variation, but future grids will require long-term storage. Both the technology and business models of BESS will need to change as grids develop, and if not, decarbonisation goals will slip beyond reach.
BESS revenue models must evolve
Solar and wind generation are inherently variable. Daily solar fluctuations follow predictable output patterns while wind patterns shift unpredictably, but these sources typically complement each other – wind peaks during winter months and in the night when solar output is low. Despite this natural pairing, extended periods of cloud cover combined with low wind conditions create a genuine challenge for a grid transitioning to renewable dominance.
Li-ion batteries (LIB) have emerged as the default storage technology to address renewable output variability, driven by falling costs and high energy density. However, their primary use today focuses on intra-day variations (i.e. where batteries store excess, low-cost midday energy and release it during higher-priced evening peak hours).
LIB systems can provide four to six hours of utility-scale storage (n.b. defined as discharge duration at maximum power), with future projects reaching up to eight hours, making them well suited to intra-day price arbitrage. This leaves longer-term fluctuations unresolved, and while an LIB could discharge at much lower ratings to extend storage duration, they are not technologically suited to holding charge for long periods.
The performance characteristics of LIB are discussed fully in the CRU Energy Storage Technology and Cost Service.
Analysis of actual solar and wind generation profiles by country shows intra-day fluctuations often represent only a minority of storage requirements in a renewables-heavy grid – much longer-term storage will be needed. Thus, as renewable penetration increases, the business models adopted by BESS (i.e. how they operate and how they are recompensed) will need to change. Long-term storage will also require different technological solutions and different revenue mechanisms.
Long-term storage becomes critical as grids decarbonise
To quantify storage requirements, we group storage requirements by the highest cycle frequency – daily (i.e. up to 365 cycles/y), weekly (i.e. <52 cycles/y) and monthly (i.e. <12 cycles/y). In reality, there will be a continuum of cycle frequencies associated with the totality of storage on a renewable-heavy grid. However, this grouping provides one means for quantifying requirements across different time frames. In populating these categories, we have made sure to remove overlapping storage requirements.
Climate and geography dictate the optimal renewable mix and storage requirements that provide the lowest cost-power. For example, Saudi Arabia has an excellent solar resource with minimal seasonal variation with consistent daily sunlight making it ideal for a solar-dominated power grid.
Modelling a theoretical 100% solar grid – using Saudi Arabia – shows daily storage requirements account for ~55% of total storage needs, broadly reflecting intra-day solar fluctuations. In this instance, LIBs provide the obvious storage solution, given the high cycle frequency in that segment and cost of the technology. However, even in Saudi Arabia, 45% of storage requirements will need to operate at weekly and monthly cycle frequencies to deal with the slight seasonality and the occasional day of poor solar output (see chart below).
For weekly cycling, LIBs could potentially still be used. However, they are not suited to sitting at full charge for long periods with such an operation reducing the lifetime of the battery. This issue becomes more relevant for the monthly storage category that, though only at ~22% of overall requirements, still represents a significant share of overall storage capacity and will require a different technological solution, such as vanadium redox or iron air batteries that are more costly.
Northern Europe (n.b. here modelled using Germany) paints a very different picture. The higher latitude leads to much greater seasonal variation in solar output, so a solar-only grid – shown here as purely an illustrative and comparative example – requires massive capacity overbuilding and correspondingly enormous battery storage needs to achieve lowest overall costs (n.b. costs that remain far higher than for Saudi Arabia). The chart below shows the overall storage requirements are much higher and more heavily weighted towards long-term needs.
Importantly, the capacity of battery needed to deal with daily variations in Germany is approximately the same as that needed in Saudi Arabia – the solar output of both follows a similar daily profile. However, Germany requires significant additional storage capacity to deal with seasonal variations in solar that are much greater in amplitude than for Saudi Arabia. LIB technology would not be able to provide storage with these low-cycle-frequency characteristics.
Mixed renewable grids need less storage than solar-only
An alternative perspective combines solar and wind, the output of which complement each other to minimise both daily and seasonal fluctuations. CRU analysis on grids concludes that a grid with >95% of power from renewables faces very high costs due to the need for extreme overbuilding of renewable capacity and large storage deployment. As a result, we also model a ‘mostly decarbonised’ grid as one with 95% renewables paired with 5% fossil energy, which significantly reduces the need for storage as flexible fossil power can absorb the more extreme variability.
In Germany, even with a mixed renewables grid – though the absolute capacity of daily storage is similar to that modelled for a solar grid – daily storage remains a relatively small component of overall storage requirements.
The chart above also shows that, with dispatchable fossil-based power available, both overall storage capacity and long-duration storage requirements fall. Flexible fossil-based power is a replacement for long-duration storage needs as it is primarily used for balancing seasonal, rather than daily, variations in output. Other fuels options such as hydrogen, or biofuels, could be used, but as fossil fuels are already widely used at scale, they would be the likely candidate to deal with the long-duration issues, at least in the short-term.
Further, we also introduced demand management in our modelling, allowing demand to fall by up to 30% within an hourly period in response to variability in renewables output.
Introducing demand management further reduces the need for storage but fails to address the core issue. Long-duration storage requirements persist because seasonal variations cannot be managed through hourly demand flexibility, which in the real world would mean imposing unrealistic constraints, such as preventing heating or lighting during winter. Demand management primarily impacts on daily variability, for which a viable solution – LIBs – already exists, but leaves the structural challenge of weekly and monthly storage unresolved.
Europe faces an uncomfortable reality
LIBs efficiently manage daily variability in power output, with regular daily charge/discharge cycles, but are technically unsuited for long-term storage that requires only a handful of charge/discharge cycles per year. Fossil fuels remain the most effective tool for mitigating long-duration storage needs but contradict decarbonisation objectives. Clean alternatives for mitigating long-term storage, such as hydrogen, remain underdeveloped and commercially unproven.
Without viable solutions, achieving grid decarbonisation will prove substantially more costly than current projections suggest, which may constrain decarbonisation objectives. This will likely lead to a mostly decarbonised grid, with fossil fuels remaining a significant minority to help deal with renewable variability, as cost pressures prevent a fully decarbonised grid. However, the desire to decarbonise does open the door to those technological solutions that can overcome LIB deficiencies in the long duration segment, but how these systems are recompensed will need to be different from BESS business models adopted today.
If you want to hear more how battery storage needs and business models will be shaped by decarbonisation, contact us and we’ll be happy to discuss our work.
