7 System value: Making sense

The increasing penetration of low-carbon generation capacity requires more power system flexibility. LCOE for generation or LCOS for electricity storage technologies are an intuitive metric for technology-specific cost comparisons. From a system perspective, however, both metrics are ambiguous, because they do not account for output variability or the impact of a technology’s operation on the electricity system in terms of reliability and operability.¹ The concept of system value determines the value of a technology to the power system as a whole as the difference in total system cost (TSC) caused by the deployment of a technology.² The concept therefore explicitly accounts for lifetime cost and the impact on power system reliability and operability, but it requires comprehensive energy system models to determine this value. The value itself can be given as the absolute difference in TSC (%), normalized per annual energy demand (USD/MWh_el), normalized per installed capacity of the technology (USD/kWh_cap), or normalized and annuitized per installed capacity (USD/kW-year).

A range of studies analyse the TSC of low-carbon power systems with variable renewable and flexibility technologies compared to systems with conventional, dispatchable generators. For the US, a detailed grid simulation model of the balancing areas in Colorado and Missouri found a system value of 145 USD₂₀₁₁/kW-year for electricity storage with 8 hours discharge duration, highlighting that the reduction of operational costs is more significant when the device provides capacity (i.e. frequency regulation) rather than energy services (i.e. following reserve).³ These results are in line with a similar model for the Texas power system, which found system values of 55–85, 120–200, and 160–270 USD₂₀₁₇/kW-year for 1, 4, and 8 hour discharge duration respectively.⁴ More broadly, a capacity expansion model of the same power system identified a 7 to 12% reduction in electricity generation investment and operation cost for 90% emission reduction as a result of electricity storage deployment. Cost savings were achieved through increased utilization of installed resources and greater penetration of lowest cost low-carbon resources.⁵ This translated to a value range of 286–572 and 103–257 USD₂₀₁₆/kWh of installed electricity storage capacity for the first 10 GW of a 2 or 10 hour discharge duration technology respectively.

In a set of European power supply scenarios, similar TSC was found for a system with 0% variable renewable electricity and one with 85% combined with electricity storage power capacity at 23% of peak demand.⁶ An integrated assessment model of 24 European countries quantified the system value of electricity storage as a 3–5 USD₂₀₁₆/MWh reduction in the integration cost of variable renewable electricity.⁷ A more simplified power system model based on long-term meteorological and load data and a 90% share of variable renewable electricity found long-term storage with an energy capacity equivalent to 168 average load hours to reduce total system cost by 10%, while more efficient short-term storage equivalent to 4 load hours achieves 20%.⁸

Table 7.1 lists studies that model Great Britain’s power system up to 2050 and include electricity storage. The penetration of variable renewables ranges from 0 to 87% of annual electricity demand and 0 to 91% generation capacity. Electricity storage capacity is included at 0.004 to 0.027% of annual demand (energy) or 3 to 57% of peak demand (power). The storage technologies modelled range from one generic proxy for all technologies to a full suite of seven different technologies. While all studies consider storage and interconnection, other flexibility options like demand-side response (DSR) and hydropower are not always included.

In terms of system value, BloombergNEF finds a 2% reduction in total system cost by 2030 in a scenario where an additional 4.8 GW of electricity storage capacity is deployed.¹⁰ This translates to a reduction of 0.7 GBP/MWh produced in the system. In comparing two scenarios with and without electricity storage, Heuberger et al. identify a reduction of 15%.² However, it is highlighted that the first GW of storage already leads to a reduction of 13%, thereby putting the result in line with the previous study (15% - 13% = 2%). The analyses by the Committee on Climate Change and the Carbon Trust determine annual TSC savings for a power system with a carbon intensity of 100g_CO2/kWh with flexibility technologies at GBP 1.4–2.4 or 3–3.8 billion in 2030 compared to no flexibility options. While the former refers to net savings, which include investment cost of the flexibility technologies, the latter uses gross savings, which does not.^11,12 At 90% VRE penetration, the value of electricity storage specifically is identified by Pfenninger et al. at 50–130 GBP/MWh_el system-wide electricity cost (32–35% of respective TSC), when adding electricity storage at a cost of 350 GBP/kWh_cap to a range of scenarios.¹⁶

Not all studies explicitly quantify the system value of electricity storage. System value originates from the ability of storage to increase the utilization of power system assets like variable or inflexible generators and thereby increase their penetration.⁵ Therefore, some studies only explore this capability without quantifying its financial value.

Figure 7.1 compares the findings of 30 studies across the US, EU, Germany, and Great Britain (GB), regarding the required electricity storage energy capacity and power capacity in low-carbon power systems with increasing VRE shares. The energy capacity and power capacity requirements are displayed relative to annual electricity and peak power demand respectively. Most studies appear to agree that for up to a VRE penetration of 50%, a power system requires less than 0.02% energy storage capacity and 20% power capacity. Taking Great Britain as an example with ~50 GW peak and 300 TWh annual demand, this would amount to 10 GW and 60 GWh of electricity storage capacity.

Storage requirements increase exponentially at higher levels of VRE penetration. Moving to 80% and 90% penetration, the energy capacity requirement increases to 0.03–0.1% and 0.05–0.2% respectively (60–300 GWh and 150–600 GWh for Great Britain’s power system). Power capacity requirements increase to 20–50% and 25–75% (10–25 GW and 12.5–37.5 GW). There is substantial variation between the findings of different studies, especially in terms of energy storage capacity, as the studies in Figure 7.1 cover several geographies and make their own varied assumptions about the mix of technologies which provide flexibility (i.e. storage vs interconnection, flexible generation, and DSR).

The trends used to summarize the results in Figure 7.1 suggest that both electricity storage power and energy capacity requirements increase exponentially with the penetration of VRE, as has been highlighted in other studies.¹⁹ Incorporating low shares of VRE is relatively easy as their variability can be accommodated by slight alterations in the dispatch of conventional power stations. As VRE share increases further it becomes increasingly difficult to manage the balance between supply and demand. Exponentially increasing power capacity is needed to consume the increasing amounts of excess renewable energy for later discharge in most cost-efficient low-carbon energy systems. The combined impact of additional variable power capacity and the time it is supposed to generate electricity means that energy capacity requirements increase at an even higher exponential rate to ensure sufficient electricity is available at all times. Not only is more backup power needed but it is also needed for longer periods of time.

Selected studies argue that energy storage power capacity increases linearly rather than exponentially with increasing variable renewable energy penetration.²⁰ That conclusion is drawn from 15 independent studies (compared to 30 studies in this book) which show notable scatter around the trend, so rather than the exact functional form of the relationship, the important conclusion is that other studies support the insight that the rate at which storage power capacity increases is smaller than for energy capacity.

This section analyses electricity storage power capacity and energy capacity requirements for Great Britain’s (GB) power system as a case in point. In this power system, nuclear is projected to play a significant role. The same is true for other power markets (e.g. France, China). Given its economic incentive for constant power output due to low fuel and high investment cost, nuclear’s impact as ‘inflexible’ technology on flexibility requirements should be considered as well, similar to variable solar and wind.

Figure 7.3 displays the electricity storage requirements as modelled by various studies for low-carbon scenarios of the future GB power system (see Table 7.1). The required absolute storage power capacity in a system with up to 90% wind, solar, and nuclear generation capacity could remain at the 4 GW installed in 2021 or increase to 35 GW (panel a). A similarly wide range is observed for electricity storage energy capacity. At 90% penetration of wind, solar, and nuclear energy generation (relative to total annual electricity demand), required storage capacity could be as low as today at 30 GWh or up to 140 GWh (panel b).

The electricity storage requirement range becomes more defined when accounting for differing peak power and annual electricity demand assumptions. The current power system at 48% penetration of wind, solar, and nuclear power capacity has 8% electricity storage power capacity relative to peak demand. According to the studies, it could range from 5–20% at 60% and 5–40% at 80% penetration (panel c). These values are equivalent to 2.5–10 GW (60%) and 2.5–20 GW (80%) electricity storage power capacity at the peak demand of 50 GW. Similarly, electricity storage energy capacity relative to annual electricity demand could remain at the current level of 0.01% or increase to 0.015% or 0.025% at 60% or 80% wind, solar, and nuclear energy penetration respectively, depending on the scenario (panel d).

Both of these results for the required electricity storage power capacity and energy capacity in GB are at the lower end of the ranges identified by the wider selection of studies and geographies in the previous section.

The shaded areas in Figure 7.3 identify possible maximum and minimum deployment levels for electricity storage subject to the assumptions in the various studies and scenarios. They identify the possible range of electricity storage capacity requirements to enable low-carbon power systems with up to 90% wind, solar, and nuclear power and energy share in Great Britain’s power system.

However, the wide ranges also highlight that low-carbon power systems do not necessarily depend on electricity storage; rather, it can be a valuable enabler under certain conditions. This insight is supported by the fact that the observed variations in electricity storage requirements are not study-specific but vary by individual scenario. For example, Pfenninger et al. feature one scenario with 15 GW storage (25% relative to peak power demand) at 85% wind, solar, and nuclear power share, compared to 3 GW (5% relative to peak power demand) in all other scenarios.¹⁶ In these other scenarios alternative flexibility technologies like interconnectors, demand-side response, gas (open cycle), hydropower, and oil and diesel generators balance variable and inflexible generation with demand.

A more comprehensive approach to assessing the system value potential of electricity storage in integrating low-carbon electricity is to analyse the overall flexibility capacity requirements, regardless of which technology provides them (see Figure 7.4). It shows that up to 40% wind, solar, and nuclear power share, less than 20% flexibility capacity relative to annual peak power demand is required. This increases to a range of 40–100% above 80% wind, solar, and nuclear power. These findings are in line with the findings in Figure 7.1.

Figure 7.4 b shows that the share of electricity storage has no impact on the required flexibility capacity. This confirms that flexibility technologies can be used relatively interchangeably. The analysis therefore shows the maximum theoretical potential for electricity storage. The practical share of each technology is determined by technology constraints (e.g. CO₂ emissions for flexible generation, limited interconnection possibilities) and their economic market value (e.g. electricity storage vs demand-side response).

This analysis reveals three insights. First, the requirement for flexibility power capacity appears to increase linearly with increasing shares of wind, solar, and nuclear power capacity. Figure 7.5 a displays linear regression trendlines as visual guides for studies with more than two data points and low variability.

Second, there appears to be a flexibility baseline at 20% capacity of peak demand. This holds from a wind, solar, and nuclear penetration of 0% up to 40% in all studies. It indicates that nearly half of power system electricity can come from relatively inflexible or variable sources before there are additional needs for flexibility.

Third, there seem to be two schools of thought in terms of modelling flexibility capacity requirements. This becomes evident when comparing studies from industry and government with those from academics (see Figure 7.5 b). Academic studies suggest half as much impact from adding low-carbon generation: an extra 0.8% for each 1% share of wind, solar and nuclear, versus an extra 1.6% from industry and government studies. This equates to an average of 60% versus 100% flexibility capacity at 90% VRE and nuclear share. This is likely the result of modelling different system margins for firm reliable capacity (i.e. all capacity except wind and solar). For most academic studies, the margin is less than 20% above peak demand (see Figure 7.6). Hence, this is the optimistic school of thought. Industry and government studies model more than 20%, making those the conservative school of thought. Section 7.6 assesses possible motivations for both approaches.

The two schools of thought can be used for two different approaches to plan low-carbon power systems. The more conservative one suggests that flexibility capacity is only needed once wind, solar, and nuclear make up 30% of the generation portfolio. It will then increase by 1.66% relative to peak power demand with each additional 1% of low-carbon power capacity (see Figure 7.5 b). According to this approach, a power system based only on wind, solar, and nuclear power would need flexibility capacity at 115% of peak demand.

The less conservative approach in academic studies suggests that no flexibility capacity is needed below 17% wind, solar, and nuclear power penetration. It will increase by 0.81% of peak power demand for each additional 1% low-carbon power capacity in the power mix. According to this approach, a power system based only on wind, solar, and nuclear power would require flexibility power capacity at 65% of peak power demand.

The two approaches outlined in the previous section can be useful in planning low-carbon power systems to assess flexibility power capacity requirements that could be fulfilled with electricity storage. Figure 7.7 uses these approaches in a thought experiment on the amount of flexibility required globally if the power generation mix changes in line with projections made in the IPCC 1.5 °C report to keep global average temperature increase below 2 °C.^32,–34

In 2020, wind, solar, and nuclear power penetration relative to total bulk generation capacity was just over 20%, and would require nearly no flexibility capacity. The 1,300 GW hydro and oil-based capacity and around ~400 GW electricity storage, demand-side response and interconnection that is deployed amounts to nearly 40% of the global noncoincident peak power demand of 4,800 GW (see Chapter 8 for details).

As the power penetration of wind, solar, and nuclear increases to 76% by 2050, the flexibility capacity is projected to decrease from 40% to 20%. This is because peak demand increases faster than projected additions for hydro- and oil-based generation, the only flexibility technologies explicitly listed in the database behind the IPCC reports.^33,34 This reveals an increasing gap between flexibility requirements and installed flexibility capacity from the early 2030s (conservative) or late 2030s (optimistic approach) onwards. The conservative approach requires additional 400 GW flexibility already by 2035. For 2050, the optimistic approach suggests 2,100 GW additional flexibility capacity and the conservative another 2,600 GW on top. At even higher penetration of low-carbon generators, Jacobson et al.’s roadmap for a 100% renewable energy system in 139 countries in 2050 finds flexibility capacity requirements at 60% of peak power demand, which is closely aligned to the suggestion of the optimistic approach.³⁵

This ‘thought experiment’ shows that, notwithstanding the limitations discussed in section 7.6, the outlined approaches can be used for high-level approximations of maximum and minimum flexibility power capacity requirements for low-carbon power systems. These two approaches outline the theoretical need for electricity storage, form the basis to assess its total financial system value, and can guide future power system planning to ensure sufficient flexibility capacity.

Sections 7.3 and 7.4 analysed the potential electricity storage power capacity requirements to balance increasing shares of variable and/or inflexible low-carbon power supply. However, Figure 7.1 also shows that the required energy capacity increases with rising shares of low-carbon power supply, and that it increases exponentially rather than linearly.

Two perspectives are important when determining the required energy storage capacities:³⁹

There is a wide range of studies that analyse the need for energy storage from both perspectives. Many of these studies implement very detailed assumptions on technology and market-specific parameters. The priority of this book is, however, to promote a fundamental understanding on the requirements for energy storage to decarbonize the electricity and overall energy sectors. The following analysis is therefore intentionally kept simple to focus on fundamental insights that are valid across energy storage technologies and power markets.

This section systematically investigates the impact of different electricity storage energy capacities and efficiencies on integrating low-carbon electricity generation. It focuses on variable generation from wind and solar power and uses Great Britain’s power system as an example. The analysis is based on meteorological wind and solar data, assumes that electricity can flow unconstrained from generation to consumption and that all generation capacity beyond wind and solar is fully flexible.

An initial step is to identify the ideal ratio of wind and solar in the generation mix that best matches hourly demand to minimize the need for electricity storage. Hourly wind and solar generation data are scaled to meet a certain share of total electricity demand overall. Hourly electricity demand is then subtracted from the scaled wind and solar generation. This reveals how much of this generation actually meets demand and is consumed, rather than curtailed.

Figure 7.9 shows that a wind:solar ratio of 85%:15% best coincides with Great Britain’s hourly electricity demand pattern. If wind and solar generate 100% of total electricity demand at this ratio, they actually provide 78% of hourly demand. In contrast, if the wind:solar ratio was 0%:100%, only 41% could be consumed. Other studies confirm an ideal wind:solar ratio of 85%:15% for Great Britain.^43,44

The analysis also reveals that below 25% wind and solar generation there is only minimal mismatch (< 1%) with hourly demand for any wind:solar ratio. For the ideal wind:solar ratio this value increases to 50%. There would be no need for electricity storage to integrate excess wind and solar generation below these penetration levels.

The next step is to model the ability of storage energy capacity to integrate excess solar and wind generation. Figure 7.10 shows the results for different sizes of 100% efficient electricity storage. The capacities are given relative to the average hourly electricity demand (average load hours) and in absolute terms (TWh). Assuming 20% wind and solar overcapacity, such that total generation amounts to 120% of total electricity demand, 84% could be consumed without any storage (compared to 78% without overcapacity in Figure 7.9). Storage systems that could supply average demand for four hours, a day, a week or a month would increase that share to 88%, 94%, 99%, or 100% respectively.

A zero-carbon electricity system could be realized with ~110% solar and wind penetration and an electricity storage energy capacity of 27 TWh (average demand for one month). Alternatively, ~140% penetration and 6 TWh (one week) would suffice. This shows how wind and solar overcapacity can reduce the need for electricity storage energy capacity.

However, these values are significantly lower than the 67 TWh or 63 TWh identified in other studies.^43,44 The reason is that electricity storage systems are not 100% efficient. Those studies assume a round-trip efficiency of 40–45% for long-duration hydrogen storage. Applying a 40% efficiency to the 27 TWh would increase the required energy storage capacity to 67 TWh, putting this finding in line with those from other studies.

Figure 7.11 shows the impact of integrating wind and solar generation for energy storage capacities of different sizes and different round-trip efficiencies (RT). The three storage types could be classified as:

The analysis reveals that up to 80% wind and solar generation, the small, short-duration storage is as effective as medium- and long-duration storage (only 1.5% difference in the amount of wind and solar generation that can be used at the hourly level). Up to 90% penetration, medium-sized and medium-duration storage is as effective as long-duration storage (only 1% difference). However, to fully meet electricity demand with wind and solar generation, large, long-duration storage is needed along with over-building renewables. In this example, it would amount to ~27 TWh at 40% round-trip efficiency and 40% wind and solar overcapacity. Such large-scale electricity storage systems do not yet exist.

Table 7.2 summarizes the key insights on the role of electricity storage in integrating wind and solar generation for Great Britain’s power system and other major power markets.

The key insights are:

These insights are in line with other studies and confirm the findings from Figure 7.1 on required electricity storage energy capacity as shown in Table 7.3.^39,43

This analysis is kept deliberately simple and high-level to transparently derive fundamental insights that hold true across power markets and scenarios. Further detail is provided in other studies, for example on:^8,43,–45

The previous section highlighted the need to deploy TWh-scale electricity storage to enable 100% wind and solar supplied electricity sectors. However, the challenge becomes even greater when expanding the perspective to the entire energy sector.

The scale of this challenge can be understood by looking at current strategic energy reserves in fossil fuels. For example, the US has energy storage reserves of nearly 5,000 TWh_calorific (based on the heating value of the fuels) for petroleum, crude oil, motor fuels, heating and other oils, and natural gas. Assuming these fuels can be converted to electricity with an efficiency of 40%, this would amount to 2,000 TWh_electric.⁴⁸ This is several orders of magnitudes larger than the electricity storage capacities identified for decarbonization of the electricity sector. Not all of these strategic reserves are for use in the energy sector; however, those that are used as industry feedstocks may also have to be replaced with electricity-derived chemical fuels like green hydrogen in future.

For a more detailed analysis on energy storage requirements to decarbonize the energy sector, Table 7.4 and Figure 7.12 focus on natural gas storage in the EU and the US. They confirm the significant storage volumes required (> 1,000 TWh_calorific) and show the distinct seasonality in charge and discharge of the storage volumes (once per year), in line with the regional heating periods. The respective maximum charge/discharge capacities reveal that the gas storage systems have minimum discharge durations in the region of 2-3 months (energy to power ratio of > 1,300 hours).

These facts show the sheer scale of energy storage that societies rely upon. The ‘equivalent’ amount of electricity storage required could be notably lower though, as the technologies for end-use conversion are more efficient. If heating is electrified via heat pumps, the amount of storage needed for heating demand could be a factor of 3 to 4 lower due to their high coefficient of performance (i.e. 1 unit of electricity is converted to 3 to 4 units of heat).⁵² However, this still leaves a requirement for hundreds of TWh_electric. A similar analysis could be conducted for energy storage reserves for transport, yielding similar orders of magnitude storage requirement.

This cursory analysis provides a snapshot of the challenge with decarbonizing the whole energy sector. The significant volumes of fossil fuel storage systems need to be converted or replaced with systems that store low-carbon based fuels/electricity. More importantly, these stores are only charged and discharged once per year, which also creates an economic challenge.

The analysis revealed a more conservative approach in industry and government than in academia for planning the future low-carbon power system. It should be noted that the academic studies focus on accurate representation of wind and solar variability using resource data from multiple years and high spatial and temporal resolution.^14,16,–18 Academic studies may also consider improvements in ancillary service provision which allows more services to be provided with less capacity.^11,53,54 Both rationales may justify system adequacy without the need for excessive capacity margins. However, temporal resolution is always above one hour, which may underestimate flexibility needs for short-term system balancing. On the other side, industry studies were conducted by direct stakeholders of the power system like the system operator, National Grid. This could mean that these studies use more realistic assumptions due to better industry insights and/or account for larger safety margins due to real-world liabilities of the stakeholders.

The role of nuclear in low-carbon power systems is highly debated. One reason is that baseload nuclear power increases the ability to meet peak power demand, which should reduce demand for flexible power capacity. However, the investigated studies optimize for most cost-efficient energy systems. Balancing renewable supply and consumer demand cost-effectively requires the ability to quickly increase or reduce power output in light of the variability of renewables and demand, which is not technically feasible or would make nuclear uneconomic.⁵⁵ Otherwise, there may be situations where renewables meet demand and nuclear produces excess electricity. Naturally, renewable generation could be curtailed in these situations and all studies assume this to a certain degree of it, but excessive curtailment will not yield cost-optimal solutions.

This contrast between providing for peak power demand but not adjusting power output reflects the wider debate about the future role of nuclear in low-carbon power systems.^56,57 Recent findings suggest value in a limited amount of nuclear to decarbonize power systems with low overall flexibility, but highlight the preference to meet peak demand with flexibility capacity.⁵³ The more detailed analysis of the investigated studies confirms that additional nuclear capacity in the presence of variable renewables is projected to actually increase the need for flexibility capacity.

Figure 7.4 shows no difference in flexibility capacity requirement between studies with more electricity storage capacity than others. It could therefore be argued that flexibility options can be used interchangeably. However, this conclusion neglects possible technology constraints in providing flexibility at certain times and durations. For example, flexible power provision through electricity storage is limited by its discharge duration, whereas provision through interconnection is limited by the spatial correlation of weather and demand patterns.⁵⁸ These limitations have direct financial implications. For example, the UK energy markets regulator ‘de-rates’ the capacity contribution electricity storage can make in capacity market auctions based on discharge duration.⁵⁹ As of 2021 storage systems with 0.5 hours discharge duration only receive remuneration for 13% of their power capacity when bidding into the one-year-ahead capacity market, while systems above 5 hours receive 95%.⁶⁰ This de-rating is supposed to reflect the equivalent firm capacity at the time and duration of peak net demand (gross demand minus output from variable renewables).⁵⁹ However, similar limitations apply to other flexibility options. Interconnector capacity is de-rated at between 10% and 97%,⁶⁰ DSR at 79% and flexible generation capacity (e.g. open cycle gas turbine (OCGT), oil, diesel, hydro) at 91–95% based on historical station availability.

For the present analysis this has two implications. First, the modelled flexibility capacity appears to refer to equivalent firm capacity, because flexibility options are not de-rated. Therefore, absolute flexibility capacity requirements may be higher than identified here. Second, the low de-rating of electricity storage with more than 5 hours discharge duration (95%) implies that this flexibility option is considered to have highest system value in meeting peak demand for the GB power system with 40–50% of solar, wind and nuclear energy share.

These cursory analyses of the required energy storage capacities to decarbonize the electricity and overall energy sectors, reveal the TWh-challenge of converting existing fossil fuel or deploying new storage systems. However, the economic case for these long-duration storage systems may be the hardest challenge to overcome. Chapter 5 shows lifetime cost of > 500 USD/MWh_electric for an exemplary seasonal electricity storage system by 2030-40. The European gas network discharges up to 800 TWh over winter, which equates to approximately 320 TWh of electricity. Given this cost of storage, society would have to pay USD 160 billion per year to move electricity between summer surplus and winter shortfall, which will be challenging to justify. Policy-makers, business professionals, and academics need to study whether flexible low-carbon alternatives like flexible generation (e.g. natural gas with carbon capture and storage, geothermal generation), demand-side response or network expansion could be more feasible and more economic. Alternatively, regulatory changes and incentive mechanisms need to be established to give the required energy storage capacity a viable business model.

The normalization of flexibility requirements to peak or annual demand or overall power capacity and energy capacity are supposed to make the identified insights applicable to all power systems. In reality, however, differences in flexibility requirements are likely to result from three dimensions:

These three dimensions should serve to qualify results based on the requirements identified in this chapter to guide power system planning and ensure sufficient flexibility capacity.

The analyses on required energy storage capacities to decarbonize the electricity sector and the energy sector as a whole are intentionally kept simple. However, the key limitations that should be mentioned are:

In more detailed studies that are supposed to enable direct policy or investment decisions, these limitations must be omitted.

This chapter has looked into the value that electricity storage technologies (and flexibility technologies more generally) provide by integrating variable renewable and other inflexible low-carbon electricity generators. While every power system is different based on available renewable resources, temporal distribution of resource availability, and existing power plant portfolio, some general trends were identified that can help to derive initial estimates of electricity storage/flexibility capacity needs to integrate low-carbon energy sources. This worked example will guide you through the derivation of these needs.

A question:

The government in Germany plans to increase the share of renewable generated electricity from ~45% in 2020 to ~80% in 2030.⁶⁴ How much electricity storage will be needed to enable that?

How to answer:

First, we need to identify how much of the renewable electricity will be variable. In 2020, 32% of electricity came from wind and solar (variable) and 13% from dispatchable hydro, biomass, waste incineration, and geothermal.⁶⁵ Assuming that the share of the latter will stay constant, the share of wind and solar must increase to 67% to hit the 80% renewables target by 2030. Second, we need to identify total annual electricity and peak power demand. These stood at 570 TWh and 80 GW.^65,66

You will see that in 2020 the German power system needed an estimated ~6 GW and ~35 GWh of electricity storage capacity. Pumped hydro capacity stood at 6.3 GW and 37 GWh.⁶⁷ The capacity of large-scale battery storage was ~0.5 GW and ~0.6 GWh.⁶⁸ Thus, there was sufficient storage capacity to integrate variable renewables in 2020.

This shows that electricity storage capacity would have to increase to 17 GW (14–26 GW) and 160 GWh (range: 120–410 GWh).

You’ve done it!

Congratulations. You have derived a first-cut estimate of the electricity storage capacity required for the German power system to increase variable renewable electricity generation to 67% (and total renewable generation to 80%). You have identified that current capacity (mostly pumped hydro) must triple in terms of power capacity and more than quadruple in terms of energy capacity. Given that suitable spots for pumped hydro in this densely populated country are already taken, this additional capacity is likely to come from battery or other novel technologies used for stationary electricity storage.

However, please be aware that these values need to be taken with caution. While the studies underlying these results look at electricity storage energy capacity needs only, the power capacity needs may also be met by other flexibility technologies (see section 7.3). The need for energy capacity highly depends on the availability of alternative generation technologies (e.g. gas peaking plants) and potential imports from neighbouring countries.