Battery Engineering

This work presents an in situ diagnosis system of large capacity lithium-ion battery based on a sponge-type battery swelling sensor, which is appropriate for battery module-level integration. In addition, developed swelling sensor was applied for battery cell and module level monitoring. The sensor could monitor the battery swelling under changing state-of-charge (SOC) with various current rates (C-rates) from 0.1C to 2C. It also demonstrated early risk detection performance that can reveal the thermal runaway of the battery module at least 1500 seconds ago, by sensing the abnormal swelling.

Recent trends in solid-state battery (SSB) development and the challenges of scaling cell-level performance to pack-level performance across various architectures and target applications are discussed. Critical integration considerations for SSB cells are highlighted, including cell stacking, compression, thermal management, and the design of busbars and pack structures to address the unique mechanical and electrochemical behaviors of solid-state battery cells and systems.

This presentation explores the fundamentals of EV battery pack safety design through thermal-runaway behavior observed in LFP and NMC cells. Internal reactions, external combustion, air entrainment, ignition sources, and vented gases and electrolyte mist are treated as interacting factors governing propagation. These insights are translated into materials-informed design principles for controlling heat transfer, oxygen ingress, flammable mixtures, and pack component combinations.

A BMS must estimate the SOC of every cell in the battery pack. The physical cell property that most significantly affects the SOC estimates produced by feedback estimators such as Kalman filters is the cell’s OCV versus SOC relationship. As a cell ages, its OCV changes, which causes inaccuracies in SOC estimates if a constant OCV relationship is assumed. This talk will present a method to estimate individual electrode states of charge (eSOC) for the cell’s anode and cathode. These eSOC estimates support diagnostics and can be combined to improve accuracy of cell-scale SOC estimates as well.

This presentation introduces an optimization-based framework for finite-horizon state-of-power (SOP) computation for lithium-ion batteries using coupled electro-thermal (CET) models. The algorithm computes maximum charge/discharge power sustained over a specified horizon while satisfying operational constraints. The constant-power requirement introduces a nonlinear relation between voltage and current, resulting in a nonconvex optimization problem. To address this, the bilinear constraint is relaxed using McCormick envelopes, yielding a computationally efficient formulation for on-board applications.

This study examines how vehicle lightweighting can enable battery resizing in battery electric vehicles, reducing battery-pack mass, manufacturing emissions, cost, and critical mineral demand without sacrificing driving range. Using an integrated modeling framework that links vehicle mass, drive-cycle energy demand, battery capacity, and chemistry selection, the study compares lithium-ion, sodium-ion, and solid-state battery pathways. The findings show that battery downsizing can deliver sustainability gains while supporting more resilient EV supply chains.

State-of-charge (SOC) can play a critical role in both the likelihood and severity of lithium-ion battery failures, yet its practical application in field mitigation strategies remains underexplored. This presentation examines how SOC influences runaway initiation and propagation in automotive-relevant cells, supported by controlled abuse testing data. The findings are contextualized within real-world recall strategies, including temporary and permanent SOC limitations.

This study investigates thermal management strategies for lithium-ion batteries operating under high-stress load profiles, highlighting the influence of temperature evolution on cell aging. The results demonstrate that effective heat transfer mechanisms can slow aging kinetics and extend cell cycle life. However, the findings also reveal that even when heat is efficiently removed from the cell surface, elevated internal temperatures may persist, continuing to accelerate degradation and reduce battery longevity.

Thermal runaway and fires with Lithium-ion batteries have been a significant concern in the past few decades as the proliferation of batteries of this chemistry in various sectors has risen exponentially. At ESRI, research into various aspects of mitigating and suppressing the thermal runaway and associated fire has been a major focus. Recent research has focused on determining the efficacy of various suppressants to suppress li-ion battery fires in confined spaces.

This talk proposes a multi-timescale electricity cost optimization framework for battery energy storage systems and validates it on a real-world deployed system. The proposed approach decomposes the problem by timescale. An upper layer uses hourly model predictive control with a rolling horizon for long-term energy arbitrage, while a lower layer employs real-time control to mitigate short-term power peaks. Comprehensive validation using 12 months of real-world operational data demonstrates a 28.6% electricity cost reduction compared to no-storage operation. The framework ensures sub-500 ms computation times, achieves a modest annual battery degradation rate of 1.20%, and delivers a 5.0-year payback period.

This talk evaluates how effective these low-cost temperature measurements are for early fault detection, using a combined modlling and experimental approach. It critically compares temperature-based diagnostics with alternative sensing methods, demonstrating that despite limitations, temperature sensing remains the most viable solution for scalable, cost-constrained battery management systems.