Electricity storage refers to technologies that store electrical energy and release it on demand when it is most needed. The storage process often involves conversion of electricity to other forms of energy and back again.2 With its unique ability to absorb, store and then reinject electricity, electricity storage3 is seen as a key solution for addressing the technical challenges associated with renewables integration alongside other solutions (e.g. more flexible demand, accelerated ramping of traditional power plants). Consequently, storage is garnering increasing
interest in the power sector and is expected to play a key role in the next stages of the energy transition.

Based on recent analysis by the International Renewable Energy Agency the renewable share of global power generation is expected to grow from 25% today to 86% in 2050. The growth is especially strong for VRE technologies – mainly solar photovoltaic (PV) and wind power – with an increase from 4.5% of power generation in 2015 to around 60% in 2050. Furthermore, almost half of PV deployment could be achieved in a distributed manner in the residential and commercial sectors, in both urban and rural locations (Figure 1).

The role of electricity storage in VRE integration
Since the first quarter of the 20th century electricity storage, mainly in the form of pumped hydro, has been used to provide a wide range of grid services that support the economic, resilient and reliable operation of power systems. The great majority of global electricity storage capacity deployed up to the present day is pumped hydro due to its favourable technical and economic characteristics (IRENA, 2017a). Over the last hundred years, the electricity storage industry has continued to evolve and adapt to changing energy and operational requirements and advances in technology.

The services that electricity storage can provide depend on the point of interconnection in the power system. For example, when connected to the grid at the transmission level, electricity storage can support increasing shares of VRE (as explained above), participate in electricity market bidding to buy and sell electricity, and provide ancillary services at the various timescales relevant to technical capabilities of each technology. When connected at the distribution level, electricity storage can provide all of the above services and in addition can be used to provide power quality and reliability services at the local substation, defer distribution capacity investment, and support integration of distributed renewable energy. It can also be connected to other generation facilities, allowing for higher price capture, provision of grid services and at the same time savings on connection costs. Finally, electricity storage can be placed behind the meter (Figure 5) to support a customer in increasing PV selfconsumption, thereby reducing electricity bills (where time-of-use demand-side management schemes exist), improving power quality and reliability, and potentially
enabling participation in energy management, wholesale and ancillary services markets through aggregators.

Physical location and operational mode (coupled with generators or standalone), along with the regulatory environment and market structure under which electricity storage operates, greatly affect the type of analysis needed to estimate both system-wide and project-wide
benefits of electricity storage. These considerations are explained in more detail in Phase 3. For example, electricity storage can be operated as a standalone unit or co-located with generation facilities, e.g. solar PV and wind farms. In the case that storage is co-located with
a PV farm, rather than being a standalone unit it is an asset of a “hybrid power plant.

Utilising the system-marginal prices from Phase 3, the various services a storage project can provide can be optimised to maximise the revenue the project receives. As a result of the optimisation, the hour-to-hour (or intra-hour) dispatch of the electricity storage project and
stacking of its various revenue streams can be visualised. Figure 7 shows the type of output from storage service stacking that can be expected from Phase 4. In this illustration, the entire capacity of a 6 megawatt-hour (MWh) electricity storage facility is used to shift VRE from hours 11–14 to hours 18–21.

Figure 8 shows an example of the outcome from a project feasibility model. In this particular example, although the system benefits outweigh the costs, the monetisable benefits are less than the costs, making the project economically infeasible for the project owner. The difference between the cost and the monetisable benefit, or the economic viability gap, if greater than zero, could be due to high storage capital costs or unfavourable market mechanisms.

Using power system models to assess value and viability

In these cost-effective cases, a variety of regulatory options should be considered to ensure that costeffective projects are deployed. Policy makers and regulators can then use the results of this analysis to identify the economic viability gap and devise appropriate incentives so that projects that are seen to be worthwhile at the system level are sufficiently compensated at the project level to move forward. This is particularly relevant in the case of a liberalised market.

The weighted average competitive scores for each technology and for each case are calculated by multiplying the competitive scores, weighting and suitability matrices in Steps 1 to 3. Technologies are then ranked based on their weighted average score for a given case, with 1
being the most suitable for a specific application, 10 the least suitable. Rankings can be shown as a heat map of how suitable each technology is for each case (see Figure 17 and Figure 18). A green colour denotes most suitable technologies while red shows less suitable ones. The topranked technologies are used in the subsequent project feasibility analysis phase of the ESVF. Please note that values in this section are purely indicative, and they have to be adjusted case by case when performing the analysis depending on the system, the technologies and other specific conditions.

Marginal peaking plant cost savings Power systems are designed with enough firm capacity
to accommodate expected demand under both normal operations and contingencies. In a grid system with a growing load, the corresponding increasing peak is usually fulfilled by building new peaker capacity, the generation resources that are only utilised during peak hours. In systems with increasing proportions of VRE, peaks in the net load become higher and narrower, reducing the operating hours for peaker plants and making a business case for electricity storage with limited capacity to replace peaker plants cost-effectively. Electricity storage can potentially provide firm capacity to the system, deferring the need for new peaker plants.

With the energy and reserve prices from the system value analysis, and the optimal dispatch results from the pricetaker storage dispatch model, the revenue of the storage project can be calculated. Based on the application ranking from the storage technology mappings – stating
which technologies are most appropriate for the case – the cost side of the analysis can be determined, including CAPEX, OPEX, depreciation and taxes. The cash flow, as well as the net present value (NPV) and internal rate of return (IRR) for the project can be calculated (Figure 25).

Using power system models to assess value and viability As the proportion of VRE in power systems increases, electricity storage is becoming recognised by stakeholders as an important tool for effective VRE integration. Several examples of how electricity storage can facilitate VRE
integration are discussed in the next part of this report (Part 3), showing how early business cases are already driving deployment of storage in some jurisdictions. Depending on the primary service the electricity storage provides, however, other technologies may be capable
of meeting the same need. The cost-effectiveness of electricity storage must therefore be assessed at system level and compared against other technologies. Past research has demonstrated that stacking revenues from the variety of services that electricity storage can
provide is key to accurately accounting for the benefits of electricity storage, as well as a necessary condition for its commercial viability. The ESVF described in this report puts emphasis on the benefits (including revenue streams) electricity storage can bring both to its owners and, more importantly, to the power system.

Real-world cases of storage use in power systems

Renewable energy has advanced rapidly in recent years, driven by innovation, increased competitiveness and policy support. This has led to the increased deployment of renewable energy technologies worldwide, with their share of annual global power generation rising from 25% today to 86% in 2050 under the International Renewable Energy Agency (IRENA) Paris compliant REmap scenario In the same year about 60% of total generation comes from variable renewable energy (VRE), mainly solar photovoltaic (PV) and wind, which are characterised by variability and uncertainty.

Electricity storage systems have the potential to be a key technology for the integration of VRE due to their capability to quickly absorb, store and then reinject electricity to the grid. Because of this, electricity storage is gaining an increasing interest among stakeholders in the power sector. Policy makers therefore need to understand the value of these resources from a technology-neutral perspective. The IRENA Electricity Storage Valuation Framework (ESVF) aims to guide the development of effective electricity storage policies for the integration of
VRE generation. The ESVF shows how to value storage in the integration of variable renewable power generation. This is shown in Figure 28.

When the share of variable renewable energy (VRE) in the system is low, operating reserve requirements have traditionally been defined as a percentage of the load or as the largest contingency of the system, or in other words, the largest generating unit at that time. With this
low VRE penetration, reserves have been divided into FCR or primary reserves, FRR or secondary reserves and RR or tertiary reserves. FCR is used to stop the frequency deviation and needs to act within the first seconds after the contingency, FRR restores the frequency to its
nominal value and acts within 30 seconds and RR is used to replace the FRR and acts within 15 minutes.

In this regard, the United Kingdom system operator, National Grid, developed the EFR product, which it defines as a dynamic service where the active power changes proportionally in response to changes in system frequency. The EFR service was created specifically for energy storage and requires a response within 1 second once the frequency has crossed a threshold, which can be either ±0.05 hertz (Hz) (service 1, wide-band) or ±0.015 Hz (service 2, narrow-band). In Figure 30 the EFR service is positioned with respect to the other frequency response services in the United Kingdom.

Besides the EFR product, which is already implemented and being used in daily system operation in the United Kingdom, there are other examples of power systems with similar
products that, although not implemented yet, will encourage the participation of energy storage in reserve provision. For example, the Australian Energy Market Operator (AEMO)
has developed an FFR product. AEMO refers to it as “the delivery of a rapid active power increase or decrease by generation or load in a timeframe of two seconds or less, to correct a supply–demand imbalance and assist in managing power system frequency.

As for the value, batteries are proven to have lowered the cost of FCAS in South Australia, as shown in Figure 34. Data show that during the end of 2016 and in 2017 payments to
existing fossil fuel generators were very high, being over AUD 7 million in some six-week periods. With the installation of the Hornsdale project, this service can be provided in a
cheaper way. In 2018 the total savings in the FCAS market are estimated at AUD 40 million.

The duck curve is already prominent in California, where it first appeared. But it has also been observed in other parts of the United States, such as in the New England states (Roselund, 2018). To manage this net load curve, the grid operator needs a resource mix that can react quickly to adjust production and meet the sharp changes in net demand. In California the first ramp in an upward direction occurs in the morning, starting around 4 am.

Figure 58 shows the solar PV share in a least-cost minigrid in 2017 and in 2030, considering two types of Liion batteries (nickel manganese cobalt [NMC] and nickel cobalt aluminium [NCA]). The graph has been prepared using results from energy modelling software (HOMER Pro) and input data from IRENA’s latest cost report on storage (IRENA, 2017a). It shows that, in 2017, development projects with a 2.5% nominal discount rate had an optimal solar PV share of about 90% with either NCA or NMC batteries. Commercial projects in a low-risk context (10% weighted average cost of capital [WACC]) had renewable share values of 44.5% with NCA and 50.7% with NMC. The results for the optimal PV share in mini-grids in a riskier context (15% WACC), typical of offgrid locations, showed a renewable fraction of only 36% using NCA and 38% using NMC batteries.

Storage deployment in an off-grid context There has been a rapidly increasing interest in deploying storage solutions in off-grid contexts, especially in minigrids that are located in rural areas where there is no access to the electrical grid or on islands that rely on expensive and polluting diesel generation. This has been driven by the need to accommodate increasing amounts of solar PV, and to a lesser extent wind, to provide electricity access or displace diesel generation.

The role of aggregators and the value they can provide to BTM storage should be noted. Aggregators are new market participants that operate a virtual power plant, which is an aggregation of dispersed distributed energy resources with the aim of enabling these small energy sources to provide services to the grid. Figure 63 is an overview of how an aggregator works.

Aggregators allow enhanced participation of BTM storage in the different electricity markets, help decrease the marginal cost of power and optimise investment in power system infrastructure; however, they require a proper regulatory framework and advance metering
infrastructure in order to exploit their full potential.

Click to download: