Capacity Fade Estimation Through a Single Relaxation Point of Lithium-Ion Battery for Electric Vehicle Applications

Currently, lithium-ion batteries (LiBs) have found widespread applications and are gaining increasing prominence in the electric vehicle (EV) sector. The accurate estimation of the state of charge (SOC) and state of health is crucial for predicting and quantifying both the remaining EV range and battery degradation. Battery degradation is commonly associated with capacity fade or an increase in its internal resistance. This study focuses on estimating the actual battery capacity while considering cycle aging for EV applications. Various methods, such as model-based and non-model-based approaches, can be employed for this purpose. Unfortunately, the former can be highly complicated to implement in a battery management system, while the latter require data collection, such as performing a complete charge/discharge cycle or analyzing the incremental capacity and differential voltage. These requirements are not feasible for EV applications because they rely on estimation rather than modeling of the open-circuit voltage (OCV) curve. To address this issue, the literature suggests methods for reconstructing the entire OCV-SOC curve from only a few fragments of the curve by applying mathematical constraints such as monotonicity and shape. However, even in this case, collecting fragments of the OCV curve can be very challenging in EV applications. In light of the above, this work proposes and analyzes the feasibility of estimating the actual battery capacity by reconstructing the entire OCV curve, combining a model-based approach with a single experimental OCV point. Specifically, this is achieved through a constraint related to the differential voltage, which remains constant during aging. The proposed method was experimentally validated, yielding promising results.