Three 100 MW-class battery storage case studies show frequency-regulation revenue and policy financing drive deployment, while thermal-runaway fires and missing standards hold it back

The study groups BESS functions in smart grids into six categories — energy storage, synthetic inertia, voltage support, peak shaving, frequency regulation and black start — and argues that in grids dominated by inverter-based renewables, millisecond electronic response can substitute for the mechanical inertia of rotating generators. Compared with earlier storage reviews focused on single technical parameters, it places the function list and the grid transition from centralized unidirectional flow to distributed bidirectional flow in one framework. A literature review and conceptual synthesis based on cited references and figures, offering no new measured data.

The Hornsdale Power Reserve case reports that the 100 MW/129 MWh lithium-ion plant saved consumers over $150 million during its initial two years of operation, cut frequency control ancillary services (FCAS) costs by about $116 million in 2019 alone, and can supply up to 3,000 MW.s of inertia against the 6,000 MW.s the South Australian grid needs for operational inertia. It presents the project's revenue and inertia contribution alongside specific figures, letting economic and technical value be read from the same case. Figures come from project technical disclosures and cited literature, i.e. secondary sources, without independent recalculation or sensitivity analysis.

Gateway Energy Storage (250 MW, LG Chem lithium-ion cells) had a section of its site catch fire in 2024, and Victorian Big Battery (300 MW) suffered thermal runaway from a leaking cooling system; together the incidents point to thermal runaway, toxic gases, re-ignition risk, poor visibility inside facilities and absent fire-mitigation standards. It maps two real incidents directly onto deployment barriers, moving the safety issue from technical parameters to siting, regulation and public acceptance. Based on incident technical reports and literature; descriptive case evidence without incident frequency or statistical distribution.

The study compiles a paired driver-and-barrier list: drivers include battery cost falling from about $1,400 per kWh in 2010 to about $140 per kWh in 2023, rising renewable penetration, ancillary-service and capacity market reforms, and government incentives and green-bank financing; barriers include high initial capital, market-revenue dependence, missing regulatory standards, environmental concerns over material extraction and recycling, and public acceptance. Cross-case comparison across three different geographic and regulatory settings yields a transferable driver–barrier table rather than a single-project lesson. A qualitative synthesis; case selection was limited to projects above 100 MW with accessible data, and the authors state reliance on secondary data with no experimental or simulation validation.

The study addresses readers concerned with grid-scale storage deployment, including power-system planners, storage project developers and policymakers, in settings with high renewable penetration and markets that need ancillary services such as frequency regulation and peak shaving. Its conclusions rest on three operating projects above 100 MW with accessible data, so it is better suited to understanding the drivers and barriers of large-scale lithium-ion BESS in real market and regulatory environments than to deriving technology choices or revenue forecasts for a specific project. The authors propose future directions including AI-based battery management for predictive maintenance and optimized charging, long-duration storage for multi-day renewable variability, hybrid battery–hydrogen or battery–thermal systems, and grid-forming inverters providing voltage and frequency control.

Readers should note that case data come from secondary sources and that revenue definitions and reporting years differ across projects, so cross-case comparison should be checked against original technical reports; the two fire incidents are described only in outline, without incident frequency or industry statistics; the battery cost decline and the IEA statement that battery adoption must grow roughly sevenfold for net zero by 2030 are cited from external reports whose scope and assumptions need checking; and the outlook on AI battery management, long-duration storage and hybrid storage is directional rather than validated by this study's data.