A flexible battery capacity estimation method based on partial voltage curves and polynomial fitting

Accurate capacity estimation of lithium-ion batteries is an important topic for electric vehicles. In real cases, electric vehicles are usually not charged from zero to full capacity; thus, the battery management systems (BMSs) of electric vehicles have insufficient data when evaluating battery capacity degradation. Considering the above conditions, we propose a flexible battery capacity estimation method based on partial charging voltage curves and polynomial fitting. First, the charge voltage curve is fitted with a polynomial function with respect to time. Then, the parameters of the polynomial functions are selected as health indicators (HIs) for the battery. Finally, these parameters are input into a deep learning model to estimate the battery capacity. Compared with existing data-driven methods, the proposed method requires less voltage data and is more flexible, allowing the battery charging capacity to be estimated in different phases. Batteries in different working scenarios were compared with advanced features proposed in the literature to validate the effectiveness and superiority of the proposed method.