Ing. Jan Kasper

Lithium-ion batteries (LIBs) play a pivotal role in various sectors such as transportation, aerospace, and stationary systems. Accurate estimation of their state-of-charge (SOC) is crucial for efficient utilization within battery management systems. This work presents an enhanced SOC estimation method for LIBs, leveraging both open-circuit voltage (OCV) and hysteresis models. A co-estimation architecture employing two estimators is proposed, firstly focusing on battery model parameter estimation, and secondly utilizing pseudo-OCV instead of voltage measurements as output. This modification offers enhanced accuracy, reduced reliance on extensive laboratory testing, and improved robustness, especially in applications with rapid temperature fluctuations. The proposed method is evaluated through dynamic discharge profile tests across temperature levels ranging from 5 to 45 °C. Root-mean-square errors of SOC estimation for various temperatures were improved from the baseline approach (0.0185-0.0420) down to 0.0090-0.0280 in the proposed approach, showcasing the effectiveness of incorporating hysteresis models into SOC estimation.