From minutes to hours: The economics driving long-duration battery storage

The rise of variable renewable energy in the modern grid is placing increasing focus on not only the role of storage - but also the duration of the required storage.

As traditional sources of generation like coal-fired power stations retire and are replaced by wind and solar, storage is critical for smoothing out the delivery of intermittent generation and meeting the system demand, as well as soaking up excess renewable generation which would have otherwise been curtailed due to network congestion.

These changing grid dynamics, where storage is used to manage the demand and supply balance of the system, require longer durations – storage assets that can discharge significant amounts of energy over extended periods.

However, the needs of the system are not always aligned with the economics of the system. Until recently, the capital cost of long duration (longer than two hours) battery energy storage systems (BESS) has been relatively uneconomic. But rapid advances in technology and manufacturing processes are seeing capital costs for long duration BESS tumble.

As long duration BESS become a realistic solution to addressing high-renewable penetration grids, the question turns to adequately incentivising their deployment.

Electricity markets with capacity mechanisms – where the installed capacity (MW) of assets is explicitly valued – are well suited to this task. The building of longer duration BESS can be explicitly incentivised by modifying the rules of the capacity mechanism.

But what about electricity grids without a capacity market? How is the installation of batteries incentivised in energy-only markets such as Australia's east coast National Electricity Market (NEM)?

BESS economics

In an energy-only market, there are two key drivers of BESS economics:

1. Energy arbitrage (buy low, sell high)
2. Ancillary services (rapid provision of power to support grid frequency changes)

In many markets ancillary services markets are primarily capacity markets – payments are based on the power (i.e. capacity) of the asset rather than the duration of response.

BESS participate in ancillary services markets by injecting (i.e. discharging) power into or withdrawing (charging) power from the grid to counteract sudden and unexpected changes in the system frequency.

The response duration is usually set by the individual ancillary service market eligibility requirements, but is typically in the order of seconds to minutes or tens of minutes. These rapid responses are well within the capability of shorter duration batteries, and to date, the economics of ancillary services markets have not aided in building the business case for long duration BESS.

What about energy arbitrage then?

A strong feature of energy-only markets is volatility – short duration spikes where the energy price climbs several orders of magnitude above the average price. These are the historical conditions under which peaking generation has been able to cover its cost base, and short duration BESS have been very well suited to taking advantage of these same dynamics.

Looking at deployed BESS in Electricity Reliability Council of Texas (ERCOT) and the NEM – two of the biggest energy-only markets globally – all of the operational BESS are between one and two hours in duration; in fact, the majority of installed BESS in ERCOT have been one hour until only recently.

The available incentives in the market fundamentally don’t favour long duration BESS.

In fact, in the NEM there is an identified gap between the system need for longer duration storage and the incentives of the energy-only market.

The New South Wales government and then the Federal government launched the Capacity Investment Scheme, an out-of-market contracting mechanism designed to explicitly support long duration storage (as well as large wind and solar projects). Long Term Energy Storage Agreements (LTESA) are only eligible for BESS longer than four hours in duration.

However, as solar - especially rooftop solar - continues to dominate the eastern Australian grid demand and prices in the middle of the day plunge further downwards, the flipside of this dynamic is elevated pricing during the morning and evening peaks, and in the longer term overnight.

These structural shifts in the intraday pricing patterns – where significant arbitrage opportunity exists outside of that presented by short term volatility – are leading to forecast pricing spreads that natively support four, eight or even twelve-hour batteries.

As the cost of batteries continues to fall, the overall economics of long duration BESS will continue to improve. Eventually, the requirement for out-of-market contracts should entirely fall away.