Energy storage is poised for rapid growth.
The past 18 months have witnessed an unprecedented number of announcements, many of them in the US.
In 2018 the state of New York announced a 1.5GW energy-storage target by 2025, New Jersey opted for 2GW by 2030, while Arizona expects 3GW is feasible by that date.
BloombergNEF links the fortunes of an energy market with high renewables penetration to cheaper energy storage.
In its 2018 New Energy Outlook, it projects that wind and solar will grow to almost 50% of world generation by 2050, on the back of "precipitous reductions in cost", but also the "advent of cheaper and cheaper batteries that will enable electricity to be stored and discharged to meet shifts in demand and supply".
According to European trade association Eurobat, costs of $273/kWh for lithium ion batteries are being reported by some manufacturers.
Focusing on the $/kWh for the battery alone is only helpful to a point, as these costs are just one component of a system that comprises inverter hardware, electronics and software.
Detailed cost breakdowns for battery energy-storage systems are scarce or difficult to nail down as this type of information is confidentially held among suppliers and integrators.
Differences in system design or technology, depending on the application, as well as system sizing and cost boundaries, can all vary, making comparison difficult, according to the International Renewable Energy Agency (Irena).
System and capital costs
Some system integrators, which typically buy in batteries and converters and incorporate them to work out the design and size of storage systems depending on their application, will be able to apply healthy margins for their proprietary software, which deploys advanced data analytics to exploit the dynamic functionality of energy storage within grids.
The contribution of cell costs to the total cost of an energy-storage system varies, depending on system size. In larger systems, power electronics and other costs become more relevant, according to Irena.
In such cases, cells might account for 35% of the system costs.
Since 2015 Lazard has provided levelised cost of storage (LCoS) analysis for the US market, releasing its latest study in 2017.
It analyses the observed costs and revenue streams associated with the leading energy-storage technologies and provides an overview of project returns.
The analysis offers an "apples-to-apples" basis of comparison among various technologies within a selected subset of identified use cases.
Lazard’s LCoS study incorporates capital costs for the whole energy storage system. These are sub-divided into the storage module, which includes the racking frame or cabinet and the battery management system, as well as the batteries themselves.
Balance of system components include the container and thermal management, power-conversion system components — including inverters and energy management system, and engineering, procurement and construction, which include project management, permitting costs, site preparation, construction and commissioning.
Levelised cost of storage
Lazard’s 2017 analysis features three front-of-meter use cases for energy storage: peaker replacement, distribution and microgrid.
Based on current energy-storage costs, distribution applications are viable, while peaker-replacement applications — defined as large-scale system designed to replace peaking gas turbine facilities — are potentially viable.
Microgrid use cases are not yet viable, according to the study.
In the US, regional grid operators have adapted market rules facilitating energy-storage deployment within different parts of the system.
For the peaker-replacement application analysis, Lazard applied revenue streams available to an independent power producer in the California Independent System Operator competitive wholesale market.
Revenues are derived from wholesale market settlement and local capacity resource programmes.
For distribution applications analysis, Lazard used revenue streams available to a wires utility in the New York Independent System Operator competitive wholesale market.
Revenues are derived from capital recovery in regulated rates, avoided cost to wires utility, and various avoided cost incentives.
LCoS comparisons are presented in several ways, including unsubsidised LCoS as $/MWh and capital cost comparison as $/kWh.
As more operational data is available for lithium ion battery systems, cost-range estimates are more accurate, compared with those for flow batteries, which Lazard’s study also addresses.
For peaker replacement, lithium ion battery-based storage costs are estimated at $268/MWh for 2018. For distribution they are estimated to be $261/MWh for 2018 (see charts, below).
For peak replacement applications, capital cost of lithium ion battery-based storage costs estimated for 2018 are $291/kWh. For distribution applications they are estimated to be $283kWh for 2018.
Cost reduction drivers
In a battery-storage technology analysis over five years between 2017 and 2021, Lazard’s analysis expects the capital cost of lithium ion battery-based storage to fall 36%.
OEM competition will continue to drive cost reductions. The market will also benefit from growing electric-vehicle battery production-capacity investment.
Competition among system integrators will stimulate reductions in balance of system and installation costs. Energy storage inverters are expected to follow similar price-reduction curves solar inverters have seen.
In most markets, energy-storage systems based on current costs and unsubsidised, are going after peaker replacement and frequency-regulation applications.
But these are niches compared with the gamut of potential applications, which include arbitrage, demand response, reserve and resource adequacy in wholesale markets, as well as distribution and transmission deferral and demand response on the distribution network.
For projects to be profitable, stacked revenue models are needed.
The availability in multiple revenue opportunities, and markets regulated in such a way as to facilitate combinations of different revenue streams that can be pursued together, simultaneously or in sequence, have a big part to play in determining whether energy storage can achieve return on investment.
In the US, projects that pair energy storage with solar are on the rise. One of the biggest is 50MW lithium-ion battery storage connected to a 65MW solar PV plant being built by First Solar in Arizona, due online in 2021.
Projects like these in states with high solar penetration can time-shift output to help meet the demand peak in late afternoon and early evening, which coincides with solar generation tailing off, avoiding reliance on ramping up peaker plants, which is a more expensive way to generate electricity.
Whether storage is part of solar projects being submitted in response to request for proposals depends on location and the offtaker’s load profile.
Adding storage helps to shape and firm the solar farm’s output to better match the load profile.
Solar and storage for time shifting is low-hanging fruit, compared with pairing wind and solar. Solar is an easier resource to predict than wind.
The lower level of predictability would have to be offset by an oversized storage system, and energy storage costs have not come down sufficiently for such projects to be viable.