Battery health reporting fails independent validation across manufacturers

Abstract

Battery state-of-health (SOH) reported by on-board battery management systems (BMS) is the primary metric available to electric vehicle (EV) owners and regulators, yet no study has validated its reliability across manufacturers against independent measurements. Here we show, through an epidemiological study of 1,114 EVs spanning five manufacturers and 375 days, that battery health reporting is fundamentally unreliable: real capacity differences of 12-25% exist within every model, but BMS SOH fails to track them, with correlations ranging from \r{ho} = 0.10 (non-significant) to \r{ho} = 0.62 only under restrictive filtering, while 384 vehicles do not expose SOH at all. A manufacturer-independent electrochemical marker achieves 74-89% degradation classification accuracy across all platforms without requiring BMS data, and a controlled laboratory validation on cells identical to those in the fleet confirms that partial-voltage-window charge measurements track reference capacity with \r{ho} > 0.80 across all 60 voltage windows (p < 0.001). These findings reveal a structural information asymmetry with direct implications for the EU Battery Regulation's 2027 SOH transparency mandate, California's Advanced Clean Cars (ACC) II durability requirements, warranty enforcement, used-vehicle valuation, right-to-repair legislation, and second-life battery markets.

Figures (5)

• Figure 1: Dataset overview and SOH landscape across 1,114 electric vehicles.a, Fleet composition across five manufacturers, showing vehicle counts and BMS SOH availability. b, Variance in independent dQ capacity explained ($R^2$) by BMS SOH, shown per platform. The best-performing platform (E-GMP 180S) explains 38.3% of capacity variation; most platforms show near-zero explanatory power, and MEB does not expose SOH through OBD-II.
• Figure 2: Validation of the dQ capacity measurement methodology.a, Multi-window consistency: distribution of pairwise dQ ratio CV across four non-overlapping 0.1 V/cell voltage windows. Median CV ranges from 4.3% (E-GMP 180S) to 7.1% (Niro/Kona); see text for ground truth validation results (Extended Data Table \ref{['tab:groundtruth']}). b, Schematic of the constant-current charging dQ measurement protocol.
• Figure 3: Model-dependent unreliability of BMS SOH.a--d, Scatter plots of independently measured relative capacity vs. BMS SOH for four platforms with SOH data. All data points are shown ($n$), but Spearman correlations are computed from the CC-filtered subset ($n_{\mathrm{CC}}$; Table \ref{['tab:capacity']}). Commercial vehicles (a): $\rho = 0.24$ (non-significant), with the vast majority clamped near SOH = 100%. E-GMP 192S (b): $\rho = 0.10$ (non-significant). Niro/Kona (c): $\rho = 0.17$ (non-significant). E-GMP 180S (d): $\rho = 0.62$ ($\textit{p} = 0.005$, CC-filtered $n = 19$; unfiltered $\rho = -0.18$, $n = 63$). e, Relative capacity distribution for MEB vehicles (no SOH available). The bimodal pattern likely reflects unresolved battery pack variants (different net capacity trims) that cannot be distinguished through OBD-II telemetry metadata. f, Summary of SOH--capacity correlations across platforms, with E-GMP 180S as best case ($R^2 = 38.3$%).
• Figure 4: SOC dose--response association with battery health.a, Time fraction in each SOC bin vs. health outcome across platforms. High SOC ($>80\%$) is consistently associated with worse health, while moderate SOC (30--50%) is associated with better health outcomes. b, Mileage-stratified comparison showing that at the same mileage, high-SOC-usage vehicles show $\sim$10--13% worse energy efficiency.
• Figure 5: Warranty misclassification analysis.a, BMS-reported SOH vs. dQ-based relative capacity ($n = 420$). The regression slope ($+0.02$; ideal $= 1.0$) confirms BMS SOH is independent of actual capacity. Points color-coded by platform. b, Diagnostic disagreement rate: fraction of vehicles in each bottom percentile by dQ that BMS fails to identify as worst. BMS misses 100% of the bottom 5% and 93% of the bottom 10%. c, Distribution of actual dQ-based capacity for vehicles BMS calls "healthy" (SOH $\geq 95$%). Despite uniform BMS classification, actual capacity spans 71--142%.