A Multi-Timescale Estimator for Lithium-Ion Battery State of Charge and State of Energy Estimation Using Dual H Infinity Filter

Accurate and robust estimation of state of charge (SOC) and state of energy (SOE) is critically significant for the battery management system (BMS). This requires that the estimation can be predicted reliably even under the dynamic working operation and different temperatures. This paper aims to present a multi-timescale estimator for SOC and SOE estimation using dual H infinity filter. The battery model parameter estimator and battery state estimator are executed with different timescales to address the slow time-varying on parameter and fast time-varying on state. In order to select the preferred timescale value, the timescale selection is theoretically analysed through the model accuracy and stability perspective. To assess the proposed multi-timescale H infinity filter estimator, lithium-ion batteries with different chemistries are executed on dynamic cycle at different temperatures. The results show that the SOC and SOE prediction errors are both within 1.5% under dynamic operation. The proposed algorithm is expected to be superior for practical application.