How wind and battery competition lowered balancing costs

AEMO’s 25 July 2018 review described an event on 14 January: with Hornsdale’s wind farm and battery participating in FCAS, the reported price was AUD 248/MWh, compared with a historically observed AUD 9,000/MWh. The estimated reduction in the cost of the five-hour local requirement was approximately AUD 3.5 million.

FCAS means Frequency Control Ancillary Services. The operator compares the event’s result with a historical reference. Those prices should therefore not be turned into a chart implying that the same day began at 9,000 and later fell to 248. The saving is also an operator estimate, rather than an individual battery’s invoice.

To understand the comparison, separate two questions. How much equipment exists in the electricity system? And how much of it can offer the particular service needed at that moment? A large total volume of generation does not automatically answer the second question. The service buyer needs an appropriate, available and deliverable offer.

More suitable offers mean more choice. Think of buying a specialised service: having many interested bidders is insufficient, but more qualified providers can give the customer meaningful options. This analogy explains the direction of competitive pressure. It cannot calculate a future market price without the actual supply and requirement data.

The source associates the result with participation by both the wind farm and the Hornsdale Power Reserve battery. That attribution matters. A story about the usefulness of batteries does not become more accurate by removing another participant. The interesting point is precisely that different technologies can expand the available service supply.

When considering a new project, this perspective encourages attention to other providers of the same service, not only similar batteries. A competitor may look very different and have a different main business. The market question remains the same: can its offer satisfy the same buyer’s need at the same time? Technology is part of that assessment, rather than its conclusion.

The AUD 3.5 million in this story is an estimated reduction in service procurement costs. Understanding an individual participant’s revenue would require its delivered service volumes and settlement information. These are two different accounting questions, even though both arise from the same market event. Neither figure should silently stand in for the other.

For someone exploring battery ownership, the case provides two useful perspectives. Batteries can benefit the system by creating more choice, while their owners operate in a competitive market. A project plan is therefore best understood through the service it intends to provide and the costs of delivering it. Our companion ESCRI revenue article examines those separate income lines, rather than the overall benefit to system users.