Large-Scale Energy Storage

The rapid growth of AI data centers is increasing demand for onsite BESS and UPS battery solutions to cover critical loads and improve power reliability. This presentation will compare current and emerging battery technologies including Li-ion, VRLA, Na-ion, and flow batteries across key performance metrics for data center applications. It will also discuss the current market trend, key projects, leading players, and a 10-year demand outlook.

Zinc-based batteries have gained serious traction in stationary storage providing high-power back-up power in critical infrastructure and long-duration energy storage to support the grid. This presentation will share the cutting-edge research of the DOE-funded HIDEZ national consortium to further zinc-based battery technologies leveraging support from industry to drive innovation.

The grid-scale battery energy storage market is evolving rapidly as project costs, deployment demand, and technology options reshape investment and system design decisions. This presentation examines current market realities, including cost drivers, emerging demand patterns, and the technologies influencing the next generation of BESS deployments. It will explore how evolving battery chemistries, system architectures, and performance requirements are shaping project economics, operational priorities, and long-term grid storage strategies.

AI’s surging electricity demand calls for batteries that can scale accordingly. Quino Energy’s nonflammable flow batteries use abundant raw materials, require no critical materials, and support a domestic supply chain. With pilot projects supporting AI data centers already underway, Quino is advancing commercialization through scalable manufacturing, delivering affordable energy storage for AI applications, enabling greater renewable energy adoption, and strengthening grid resilience as demand for reliable power grows.

There's no such thing as a 100% safe battery. But everyone wants a safer battery. If the cost is too high, then safety engineering is handled on systems basis, which can also be an expensive proposition. This presentation is an analysis of how much money can be saved on safety engineering if the battery cell is inherently safer. It will explore where cost savings can be found in ESS systems when safety can be re-engineered thanks to a safer cell. The presentation will also show Mana's approach to battery safety, including safety test results.

Accurate battery diagnostics and prognostics rely on accurate battery performance and degradation models. While physics-based models exist, they are challenging to parameterize. Effective methods for identifying holistic battery performance and degradation models from only ex situ data are still needed. Here, we demonstrate an empirical OCV+ECM battery model enabled by machine-learning that accurately models battery performance and degradation through its lifetime. Developed under U.S. DOE Office of Electricity, ROVI program.

As battery energy storage deployments expand, comprehensive testing is essential to ensure system safety, performance, and long-term value. This presentation will explore battery testing across the lifecycle, from initial deployment and operational performance assessment to end-of-first-life evaluation and second-life applications. It will examine how advanced diagnostics and testing approaches can support more informed decisions on battery health, reuse potential, and lifetime value.

Most public Li-ion battery studies have aged cells only to 80% remaining capacity due to historical thresholds from the automotive industry. However, depending on the application requirements, batteries may still have significant value beyond this point. This presentation will cover the effect of temperature, state-of-charge range, and discharge rate on capacity retention, energy throughput, and round-trip efficiency of commercial LFP, NCA, and NMC cells in an 8+ year cycling study.

LFP dominates grid-scale storage, but its flat voltage curve makes state of charge estimation and cell balancing harder than other battery chemistries. Across most of the operating window, cell voltage barely changes with SOC. Combined with voltage hysteresis and measurement error, this lets imbalance build up unseen until it limits usable capacity. Conventional BMSs depend on reaching top of charge to recalibrate and balance. That works for some duty cycles, but many grid applications rarely or never reach full charge. This talk explains where imbalance can and cannot be observed on the LFP curve and why one SOC and balancing strategy can't serve every application. We compare approaches for daily cycling, frequency regulation, grid forming, and backup duty cycles. These include continuous balancing during normal operation and observation windows matched to each application. We then show how these choices affect usable energy, system availability, and augmentation planning over the life of a project.