In electric vehicle (EV) applications, accurate estimation of State of Health (SOH) of lithium-ion battery packs is critical for ensuring performance reliability, operational safety and user confidence. SOH is a key parameter monitored by Battery Management System (BMS) to assess the remaining usable life of the battery and to make informed decisions regarding charging, discharging, power delivery, and maintenance scheduling. In traditional SOH estimation techniques often rely on simplistic full-cycle charge-discharge data or single-parameter tracking (such as voltage or internal resistance and other method like coulomb counting and kalman filter ), which do not account for variable field conditions such as partial and full state-of-charge usage condition, dynamic load profiles, and non-uniform aging. As a result, these methods can produce significant deviations in SOH prediction, potentially causing premature system-level shutdowns, inaccurate range estimation, and delayed detection of degradation or failure events and also customer not getting the confidence. This paper proposes an advanced methodology for SOH correction based on a segmented voltage-window-specific charge accumulation model.
This method involves monitoring the capacity accumulated within four distinct voltage windows—2.5–3.0 V (Condition A), 3.0–3.25 V (Condition B), 3.25–3.5 V (Condition C), and 3.5–3.65 V (Condition D)—during either full or partial charge cycles. The maximum cell voltage achieved during each charging cycle is used to classify the cycle and compute respective SOH segments (Soh_A, Sah_B, Soh_C, and Soh_D). These values are then compared to upper and lower threshold references derived from Life cycle data for defined charging profile to evaluate deviation. A decision matrix approach is implemented to determine whether SOH correction is required, and if so, which segment SOH value should be used to replace or update the existing SOH estimation.