TL;DR
Get backup power and energy gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
A Canary Media analysis of data from the Electric Power Research Institute and battery-industry installation figures found the annual failure rate for U.S. commercial and utility-scale batteries was more than 97% lower in 2025 than in 2016. Cumulative deployments grew 250-fold over that period, though the dataset is small and does not measure the severity of each failure.
The annual failure rate for U.S. grid batteries was more than 97% lower in 2025 than in 2016, even as cumulative installations grew 250-fold, according to a Canary Media analysis comparing industry deployment figures with failure data compiled by the Electric Power Research Institute. The finding suggests that reported battery failures have become less frequent per deployment as the technology has scaled, though it does not mean fires have been eliminated or that every incident carries the same risk.
The analysis combined EPRI’s records of commercial and grid-scale battery failures with installation data from the Solar Energy Industries Association and Benchmark Mineral Intelligence. EPRI’s failure tally includes arc-flash, fire and explosion events. The annual rate fell substantially over the period covered, and there were three years in which no known failures were recorded.
The zero-failure years do not establish that the underlying risk was zero: they mean no known failures appeared in the dataset for those years. The report also cautions that the number of recorded incidents is small, limiting the strength of statistical conclusions. The tally counts events but does not capture their scale or consequences.
Battery fires have occurred and can be difficult to extinguish using conventional methods. The source report identifies two U.S. incidents illustrating the range of outcomes: a 2019 fire near Phoenix sent four emergency responders to hospital with serious injuries, while a major blaze at California’s Moss Landing facility in January 2025 caused no reported injuries but largely destroyed the plant and led to a multi-day evacuation of nearby residents.
Falling Rates Meet Local Scrutiny
The data may help communities weigh battery proposals against concerns about fire, emergency response and nearby residents. Grid-scale batteries are increasingly being used as on-demand power resources, and the frequency of failures relative to deployment is one measure relevant to those decisions. But a falling rate does not remove the need to examine site-specific hazards, safety plans or the possible consequences of a low-frequency incident.
The report also compares battery-unit failure rates with those of oil-filled substation transformers, finding them comparable. That comparison offers a familiar reference point for grid operators, but it does not show that the equipment has identical hazards or that communities face the same consequences in every case. EPRI’s data do not provide a detailed severity-adjusted comparison.
Public debate can affect whether and where projects are built. Battery storage can support grid reliability and may place downward pressure on electricity costs; projects can also bring tax revenue to host communities. A clearer account of incident frequency could inform those discussions, while the small sample and uneven safety rules across countries argue against treating one rate as a complete measure of risk.
Incidents and Safety Changes
Battery projects have drawn public concern in part because some fires are difficult to control and because large facilities are often near communities. The Moss Landing fire in January 2025 became a prominent example in debates over new projects: the facility was largely destroyed, and residents were evacuated for multiple days, although no injuries were reported in that incident.
The EPRI report covers failures around the world, while the Canary Media analysis focused on the U.S. by pairing those records with U.S. deployment figures. That distinction matters because countries differ in their rules for energy development and safety. The source report says new safety standards took effect after the first large battery installations, while equipment and project layouts also became more sophisticated. The analysis shows the trend over time but does not isolate how much each change contributed to it.
EPRI’s Lakshmi Srinivasan, the report’s lead author, said the low number of failures is encouraging but makes the dataset difficult to analyze robustly. She also said battery projects receive public scrutiny that she does not currently see for transformers or substation construction, despite the comparable failure rates described in the report.
“There just aren’t a lot of failures, which is a good thing, but that means you don’t have a very robust dataset to analyze.”
— Lakshmi Srinivasan, lead author of the EPRI report
Limits of the Failure Records
The reported decline is based on a small number of recorded failures, and the dataset does not capture incident size or severity. It therefore cannot show from the tally alone whether the consequences of failures have changed alongside their frequency. The source material also does not provide the underlying annual counts or enough detail to independently assess how consistently incidents were reported.
The analysis combines global failure records with U.S. deployment figures to estimate the U.S. trend. Different safety rules and development practices across countries may affect how well global event records reflect U.S. conditions. The source does not establish how much of the falling rate is attributable to new standards, improved equipment, changes in project design, or other factors.
Nor does the comparison with transformers mean the two technologies have identical risk profiles. The report’s finding concerns failure rates; the available material does not provide a like-for-like comparison of fire outcomes, injuries, property damage or emergency-response demands. Future incidents and better data could change how confidently the trend can be measured.
More Data Will Test the Trend
The next step is continued tracking of battery installations and incidents as the U.S. fleet grows. A larger record could help researchers assess whether the lower annual failure rate persists and clarify the relationship between equipment, project layouts and safety practices. The source material does not identify a specific upcoming report or publication date.
For communities considering projects, the current figures provide one piece of evidence rather than a site-level safety verdict. Questions about emergency planning, facility design and local protections remain relevant, particularly given the differing outcomes of past incidents. More consistent reporting on incident frequency and severity would help residents, developers and grid operators make more informed comparisons.
Key Questions
How much did the U.S. grid battery failure rate fall?
The Canary Media analysis found that the annual rate in 2025 was more than 97% lower than in 2016. It also reported that cumulative deployments grew 250-fold over that period.
Does the data mean grid battery fires are no longer a risk?
No. Battery fires have occurred, and some have caused serious disruption or injury. The data show a lower recorded annual failure rate relative to deployment, not an end to failures or a guarantee that future projects will be incident-free.
What are the main limits of the analysis?
EPRI’s dataset contains relatively few failures, so the sample is limited. Its tally also does not account for the size or severity of each event, and the analysis does not establish which safety or design changes caused the rate to fall.
How does the battery failure rate compare with transformers?
EPRI calculated that grid battery units fail at a rate comparable to oil-filled substation transformers. That is a comparison of failure rates, not proof that the technologies have identical hazards or consequences.
What happened at Moss Landing?
A major fire at California’s Moss Landing battery facility in January 2025 caused no reported injuries but largely destroyed the plant and prompted an evacuation of the surrounding community for multiple days.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
