Large-Scale Battery Energy Storage Systems

The rapid expansion of electric mobility and stationary energy storage systems has intensified the need for accurate, real-time estimation of battery State-of-X (SOX) indicators, including State of Charge (SOC), State of Health (SOH), and Remaining Useful Life (RUL). These metrics are essential for ensuring safety, maximizing usable energy, and reducing operational costs across electric vehicle and grid-connected applications, particularly for Lithium Iron Phosphate (LFP) batteries. Although conventional machine learning and deep learning methods deliver strong predictive accuracy, their computational complexity and reliance on floating-point operations limit deployment in resource-constrained battery management systems. This work explores lightweight spiking neural network architectures that leverage event-driven, sparse computation to enable low-power, low-latency, and accurate SOX estimation suitable for both onboard and stationary energy storage systems.

This presentation introduces a deep generative transfer learning for immediate remaining capacity estimation of heterogeneous second-life batteries using short pulse tests. Without relying on historical usage data, the method synthesizes voltage responses across state-of-charge conditions and aligns heterogeneous battery domains, enabling accurate capacity assessment with minimal field data. The approach significantly reduces testing time and uncertainty, supporting scalable and low-risk deployment of second-life batteries in large-scale energy storage systems, improving confidence in reuse decisions and asset valuation. We will cover immediate capacity estimation without historical data, generative modeling and domain adaptation for data heterogeneity, and data efficient diagnostics for second-life BESS.

Sodium-ion batteries are emerging as a compelling option for grid-scale energy storage, but successful deployment depends on aligning chemistry capabilities with application-driven requirements. This talk presents Peak Energy’s cell-level approach to NFPP/HC sodium-ion development, supported by performance data from large-format cells. The discussion highlights cycling behaviour, calendar-life considerations, and safety-relevant characteristics, with a focus on the cell-level tradeoffs that enable scalable, long-life battery energy storage systems without delving into detailed materials or process design.

Lyten has been developing lithium-sulfur (Li-S) battery technology since 2018, and, following the recent acquisition of the former Northvolt assets, has expanded into giga-scale cell production.

The fraction of renewable energies in European power systems is steadily increasing, e.g., Germany already achieved approximately 60% and most of the energy is provided by the volatile sources wind and solar. A further increase of renewables requires flexible assets integrated into the grid on a large scale and batteries are highly suitable for fulfilling almost all related tasks. Additionally, energy transition results into a shift in the paradigm towards grids being operated in a safe and reliable way with power electronic assets. Thereby energy storage with grid forming inverters play a key role. This presentation provides an overview on key market developments and technology solutions to meet flexibility and grid stability at the same time.

This talk will cover Envision's updates on BESS safety, grid stability, and long-duration energy storage.

Thermal runaway is a defining safety concern for battery energy storage system (BESS) applications. Reliably interrupting the exothermic chain reaction to mitigate low-frequency, high-severity events is a critical safety objective, while minimising the likelihood of costly battery damage or lost availability from unintentional deployment is an equally important performance goal. This work describes the successful design, validation, and integration of a novel-agent direct injection fire suppression system into the Volvo Energy PU2000 BESS to achieve its targeted safety performance and support dependable, large-scale energy storage deployment.

UL9540A has been established for 11 years as the leading fire safety standard in the US, influencing global standardization and best practices — including upcoming harmonized EU standards under Article 12 of the EU Battery Regulation. Revision 5 has now been published as a draft and is expected to take effect in 2026. The new revision introduces major changes, particularly in the Large Scale Fire Test. Additionally, a specialized variant for residential applications is available under UL9540B.

NXP’s BESS 1500V reference design supports maximizing battery longevity by combining multiple technologies not only as begin-of-life datasheet specifications. High precision voltage measurements qualified for >20 years, backed with a safety certified system layout, Cybersecurity and combined with cost effective end-of-life passive or active balancing capability maximizing State of Health.

This presentation will explore the size and growth of the US stationary energy storage market. A special focus will be on datacenter battery usage and deployment.

As AI continues to advance and integrate into everyday applications, energy consumption is expected to surge. This talk will explore how Battery Energy Storage Systems (BESS) can support the rapid expansion of data centres, ensuring reliability, efficiency, and sustainability. We will discuss the benefits of behind-the-meter (BTM) storage, examining how onsite battery solutions help data centres manage peak loads, reduce costs, and enhance resiliency.