Method for Estimating the SOH of Lithium-Ion Batteries Based on the Impedance Trend

Objective: to propose a method for estimating the state of health (SOH) of lithium-ion batteries based on a low-dimensional description of the normalized impedance trend Z(p), formed from time-domain voltage and current responses under AC excitation without performing full frequency-domain electrochemical impedance spectroscopy. Methods: the initial impedance response was segmented according to current polarity, normalized in the progress coordinate p ∈ [0, 1], and averaged to obtain the impedance trend Z(p).Three diagnostic indicators were extracted from this trend: the plateau mean value P, the plateau slope magnitude S, and the end-region deviation E. This representation makes it possible to replace the full impedance-response curve with a compact set of parameters suitable for subsequent regression processing. The relationship between these features and SOH was established using a quadratic regression model, with coefficients determined by the least squares method. Results: experimental validation on LIR2032 and NCR18650B cells showed that the P, E, and S indicators reflect changes in the level, shape, and end regions of the impedance trend during battery aging. For LIR2032, R² = 0,9987 with RMSE = 1,23 % was obtained; for NCR18650B, R² = 0,9715 with RMSE = 5,19 %. Practical significance: the proposed approach is characterized by low computational complexity and interpretable features and can be used as a basis for BMS algorithms and online diagnostics of battery systems. The method can be implemented without a significant increase in computational load and adapted to different types of lithium-ion cells after model calibration.