Uncertainty Analysis for Battery Health State Based on Bat Algorithm

Online estimation of the state of health (SOH) is critical to the safe and stable operation of battery management systems (BMS), which improve energy utilization and ensure battery safety. Therefore, we propose a new algorithm—Kernel Extreme Learning Machine(SOH) based on Bat Algorithm(BA) optimization—which combines BA and SOH. At the same time, considering the existence of uncertain parameters in the data set, the universal grey system theory is used to predict the SOH. The simulation results show that the BA provides the optimal solution of hyper-parameters for the KELM, improves the precision of the model and reduces its complexity. The feasibility of this method is proved by the SOH estimation of four different types of batteries.