Robust Estimation of Battery State of Health Using Reference Voltage Trajectory
Rui Huang, Jackson Fogelquist, Xinfan Lin · 2025
Read Robust Estimation of Battery State of Health Using Reference Voltage Trajectory by Rui Huang, Jackson Fogelquist, Xinfan Lin (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-13324.txt.
Introduction
Rui Huang, Jackson Fogelquist, and Xinfan Lin present a sturdier way to gauge battery health in a paper from 2025. Knowing a battery's state of health matters for cars and grid storage. Model-based methods often need slow, gentle test cycles that real use rarely allows.
Errors in the model can also throw off the health estimate. The new method stores the voltage curve of a fresh battery under a set test pulse. Years later, the same pulse is run and the new curve is compared with the stored one. The gap between curves corrects for model error. That gives a health reading that holds up better in real use.
Catalog record
| CREATOR | Rui Huang, Jackson Fogelquist, Xinfan Lin |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-13324
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Robust Estimation of Battery State of Health Using Reference Voltage Trajectory Rui Huang, Jackson Fogelquist, Xinfan Lin Abstract Accurate estimation of state of health (SOH) is critical for battery applications. Current model-based SOH estimation methods typically rely on low C-rate constant current tests to extract health parameters like solid phase volume fraction and lithium-ion stoichiometry, which are often impractical in real-world scenarios due to time and operational constraints. Additionally, these methods are susceptible to modeling uncertainties that can significantly degrade the estimation accuracy, especially when jointly estimating multiple parameters. In this paper, we present a novel reference voltage-based method for robust battery SOH estimation. This method utilizes the voltage response of a battery under a predefined current excitation at the beginning of life (BOL) as a reference to compensate for modeling uncertainty. As the battery degrades, the same excitation is applied to generate the voltage response, which is compared with the BOL trajectory to estimate the key health parameters accurately. The current excitation is optimally designed using the Particle Swarm Optimization algorithm to maximize the information content of the target parameters. Simulation results demonstrate that our proposed method significantly improves parameter estimation accuracy under different degradation levels, compared to conventional methods relying only on direct voltage measurements. Furthermore, our method jointly estimates four key SOH parameters in only 10 minutes, making it practical for real-world battery health diagnostics, e.g., fast testing to enable battery repurposing. I INTRODUCTION Accurate monitoring of the state of health (SOH) of lithium-ion batteries is essential for ensuring their safety, reliability, and efficacy in applications, including electric vehicles (EVs) and energy storage systems [ 1 ] . Recently, there is an emerging demand for fast and accurate estimation of battery SOH to enable battery repurposing at a large scale [ 2 ] . Model-based SOH estimation [ 3 ] has been widely used due to its ability of estimating internal battery parameters directly indicating SOH. This practice involves finding the model parameters that best fit the model output with the measurement directly. The results, however, are heavily affected by the inevitable modeling uncertainties, i.e., the mismatch between the true dynamics of the battery and those captured by the model, even if the model is correctly parameterized [ 4 ] . It is noted that SOH indicators are typically associated with the parameters of the battery open-circuit voltage (OCV), such as the solid-phase active material volume fraction \varepsilon_{s} and lithium-ion stoichiometry \beta , which are directly linked to the battery’s capacity and ability to store and deliver energy. For example, in [ 5 ] , the electrode capacity and stoichiometry were estimated as SOH indicators by fitting the model output to low C-rate data (1/20 C). Schmitt et al. [ 6 ] proposed to use 1/15 C current to reconstruct OCV and estimate capacity based on charging/discharging data at rates lower than 1/4 C. Rahman et al. [ 7 ] explored the use of Constant-Current-Constant-Voltage (CC-CV) charging/discharging protocol (0.7 C) to estimate battery electrode SOH, but with additional instrumentation such as the case reference electrodes. It is seen that these methods typically rely on low C-rate constant current testing over a wide range of battery state-of-charge (SOC) to measure the battery OCV with good fidelity, resulting in an extended testing duration. However, such slow tests are rarely feasible in practical applications where timely assessment within a short testing period is necessary, and hence higher C-rate data need to be used. Under such conditions, the voltage response of the battery is significantly influenced by dynamic effects such as polarization and diffusion [ 8 ] , and deviates from the open circuit condition. It should be noted that these dynamic effects are very difficult to be modeled exactly, and the residual, which we refer to as the model uncertainty will make it extremely challenging to accurately retrieve the OCV-related parameters. In this paper, we address these challenges by proposing a reference voltage-based method that can compensate for modeling uncertainty. The key idea is to measure the voltage response of a battery under a designed current excitation at the beginning of life (BOL) to establish a reference voltage trajectory. This reference, combined with the voltage response of BOL battery model under the same input, inherently captures the modeling uncertainties under the test setup. As the battery degrades, the same excitation is applied to generate the voltage response, which is compared with the BOL trajectory to estimate the key health parameters accurately. To further enhance parameter identifiability and estimation accuracy, we design optimal current excitation profiles using Particle Swarm Optimization (PSO), which maximize the information content of the target parameters in the voltage responses. Simulation studies conducted under various degradation levels demonstrate that our proposed reference voltage method significantly improves estimation accuracy compared to conventional approaches that estimate directly from voltage measurements. Our approach enables the estimation of four key battery health parameters within a short testing time of 10 minutes. The proposed method’s enhanced accuracy, speed and ease of implementation make it particularly advantageous for practical applications, e.g., the repurposing of batteries for second-life use, where quick and accurate SOH assessment and testing are crucial. II BATTERY MODELING We use the single particle model with electrolyte dynamics (SPMe) as the model for battery SOH estimation as it strikes a good balance between accuracy and computational cost [ 9 ] . The model is a simplification of the full-order first-principle Doyle-Fuller-Newman (DFN) model, where the diffusion of lithium-ion in all solid particles across an electrode is represented by one single spherical particle, neglecting the spatial difference between different particles and assuming uniform reaction current density along the electrode thickness direction. In SPMe, the input is the current I , and the output voltage V is mainly comprised of four parts, as shown in ( 1 ), where the subscript n denotes the variables of the negative electrode (anode) and p denotes those of the positive electrode (cathode). The first one is the difference in open circuit potential (OCP) U between the two electrodes, which are nonlinear functions of the solid particle surface lithium-ion concentration c_{se} driven by solid phase diffusion. The second one is the difference between electrodes in overpotential \eta , which drives the intercalation/deintercalation of lithium at the particle surface through the Butler-Volmer equation. The third one \phi_{e} is the electrolyte potential (difference) due to the concentration gradient of lithium-ion between the two electrodes, driven by electrolyte lithium diffusion. Finally, the term IR_{l} represents the ohmic effect of the lumped battery resistance, including the contact resistance, electrolyte resistance and that of the solid-electrolyte interphase (SEI). \begin{split}&V=\\ &(U_{p}(c_{se,p})-U_{n}(c_{se,n}))+(\eta_{p}(c_{se,p})-\eta_{n}(c_{se,n}))\\ &+(\phi_{e,p}(c_{e,p})-\phi_{e,n}(c_{e,n}))-IR_{l}\end{split} (1) It is noted that under the conventional low C-rate testing schedule, the battery is close to equilibrium, that is, the overpotential \eta , electrolyte potential \phi_{e} and ohmic voltage drop IR_{l} are all negligible, and the concentration of lithium-ion in the particle is uniform. As a result, the output voltage is dominated by OCP, resulting in the battery OCP model, as shown in ( 2 ). V\approx U_{p}(c_{se,p})-U_{n}(c_{se,n}). (2) Details of the dynamic voltage components can be found in literature [ 9 , 10 ] , and are omitted here. In ( 2 ), c_{se} is the same as the