Storage

Most energy storage tech is new. Do we know how reliable it is?

Most energy storage tech is new. Do we know how reliable it is?
December 15, 2022 - Electrical engineering researcher Kurtis Buck opens the doors of the KORE POWER battery units outside the Energy Systems Integration Facility at the National Renewable Energy Laboratory (NREL). (Photo by Werner Slocum / NREL)

A good portion of energy storage technology is still relatively new as the energy industry adapts to the energy transition. While the industry should be lauded for adopting resiliency measures like energy storage, there are still gaps and little to no firm understanding of long-term reliability.

A new report from the Electric Power Research Institute (EPRI), Pathways to Improved Energy Storage Reliability, explores the challenges of assessing reliability for the large swath of storage technologies and delves into current indications from reliability data. The report also provides a framework meant to allow for more clarity in storage reliability, in addition to results from EPRI members that highlight member needs in terms of reliability and emerging policy impacts.

Challenges in energy storage

The U.S. alone has installed more than 15 GW of energy storage, the report said, but it’s still difficult to determine how reliably those systems operate. EPRI said there appear to be indications that some storage systems face issues and lower reliability when compared to legacy electric utility assets.

Historic data for “commercially oriented” storage systems typically spans less than five years, the report said, and there is not much uniformity to the structure or extent of the data. However, there is plenty of reliability assessment data on legacy utility assets, as they have more than 30 years of operational data from a “wide variety” of equipment and manufacturers, the report said.

This data provides tangibility to equipment failure rates, failure mechanisms, maintenance procedures, and a firm understanding of asset life, the report said – all of which are lacking for energy storage. Additionally, energy storage systems are deployed across multiple ownership models, including power purchase agreements (PPAs) and third-party-owned and -operated models, which often leaves operational data inaccessible to operators or offtakers.

Besides the lack of data, additional obstacles are preventing an accurate assessment of energy storage reliability. One such obstacle is the “rapid technology growth” and continued introduction of a wide variety of technology solutions and products, according to EPRI’s report. Lithium-ion is the most common storage technology today, but it alone has multiple chemistries and configurations that alter the operational characteristics and supporting systems. Alternative battery technologies have begun to emerge, such as sodium ion, but they have a virtually non-existent track record of grid operations at scale.

A new framework

To address these challenges, EPRI created an overall framework for its research that links performance analysis to reliability analysis, which is meant to inform maintenance leading practices and asset management practices, it said — a “continuous, circular effort” informed by field data and experiences.

To create the framework, EPRI relied on its own research and experience in assessing the reliability of legacy assets like generating units, transformers, and other equipment; in addition to reliability research on similar inverter-based resource (IBR) data. EPRI said the ultimate focus is on the individual components of storage systems, and their contribution to suboptimal system performance.

An early look at reliability

There’s still plenty of work to be done, but EPRI has conducted investigations of daily outage and curtained and non-operational data produced by independent system operators (ISOs). It calculates unavailability from the data collected on a daily basis, which is meant to provide more visibility into why a system is not participating in the market.

EPRI has been collecting data for more than a year from the California Independent System Operator (CAISO) and recently started collecting data from the Electric Reliability Council of Texas (ERCOT).

Based on the data it has collected so far, EPRI found that storage typically has the highest unavailability when there is an isolated “plant trouble” cause. CAISO defines plant trouble as when “plant equipment fails or is in danger of imminent failure resulting in a curtailment of dispatchable capacity.” Additionally, EPRI found that storage typically experiences almost three times the amount of plant trouble when compared to other IBRs.

Finally, when looking at the reasons for unavailability with certain systems, EPRI found that unavailability over time is not just due to plant trouble, but is also caused by metering issues, telemetry issues, and maintenance.

Read the full report here.