When project developers evaluate the potential profits for a new battery project in the NEM, one key variable being considered is the maximum allowable cycling rate of the battery. Cycling at higher rates can enable more arbitrage profits to be earned, but every MWh of energy that passes through the battery results in irreversible changes to the battery chemistry that lowers the total storage capacity over time. This reduced capacity can lead to lower profit potential in later years and additional costs to manage the accelerated degradation.
In this Chart of the week, we look at how some currently operating batteries around the NEM have been cycled over the last financial year using publicly-available SCADA data. We compare these to how batteries modelled using Cornwall Insight Australia’s Storage Investment Model (SIM) would have performed over the same period.
The SIM maximises the net operating profit for each trading day, optimising the dispatch of the battery in each market (wholesale + 8 FCAS) for each half-hour interval. Four 100 MW batteries of 1.5hr storage capacity[1] have been modelled with no cycling constraint in four NEM regions (excluding TAS) for comparison, as most of the currently operating batteries in the NEM have between 1.2 to 2hrs of storage. The monthly average cycling rates for each battery compared to the actual operations of the extant batteries, as depicted in Figure 1.
This figure shows the cycling performance for most batteries currently operating in the NEM tracking along the same trajectory as for the unconstrained modelled batteries, albeit at a lower level. The NEM batteries would be operating under various cycling constraints (which are described later in this article) that limit the ability of these batteries to exploit more than one arbitrage opportunity per day, whereas the modelled batteries have no cycling constraint imposed and can pursue all available opportunities.
Higher cycling rates were seen in mid-2022 while price conditions were elevated and supply/demand conditions were tight during the post-market suspension months. This generally allowed for both the evening and morning peak demand periods to be exploited for arbitrage revenues. As these conditions eased towards the end of 2022 and through the summer months, the arbitrage opportunity in the morning peak demand period lessened, leading to a reduction in average cycle rates during this time.
It should be noted that merchant arbitrage potential is not necessarily the main driver of the operational philosophy of the operating batteries in the NEM. For example, the Victorian Big Battery is contracted for 250 MW of network support provision for November-March each year, leaving only 50 MW active in energy trading and subsequently low cycling rates relative to the nominal capacity.
In reality, the ability to reach the theoretically optimal cycling performance that maximises arbitrage revenue will be limited by a number of factors that developers and operators must continuously keep in mind:
- The decision-making process of the battery operator (or its chosen bidding platform) on how and when to dispatch into the energy and FCAS markets
- The accuracy of pre-dispatch forecast prices, including whether all price spikes can be forecast in advance or if those forecast high prices actually materialise
- The technical performance of the battery and its capacity to run for sustained periods at higher cycling rates. Manufacturers typically impose limits on cycling duty (e.g. 1 cycle/day or 365 cycles/year) to comply with battery cell warranty conditions, limiting the propensity to chase higher revenues with more aggressive cycling
- The additional costs associated with dealing with faster or more severe battery degradation, including in ongoing operations and maintenance activities or in replacing degraded battery cells if cycling duty limits are breached.
Ultimately, battery operators must determine how to approach each of these issues and chart their own path to maximise their returns on investment for their battery project. Our in-house Storage Investment Model can provide in-depth revenue forecasting and optimisation services to help investors and developers determine the right approach for their projects. For more information, please contact enquiries@cornwall-insight.com.au.
[1] Key modelling assumptions:
- Battery registered to provide up to 50MW of Regulation FCAS and 52MW of Contingency FCAS (average recent Contingency FCAS registration of other market batteries in the NEM)
- Fixed 85% round-trip efficiency for modelled batteries, battery capacity degradation ignored
- NEM prices for 1 July 2022 – 31 May 2023 used as model inputs
