GRU Optimized by Beetle Antennae Search Algorithm for State of Health Estimation of Lithium-ion Battery

State of health (SOH) estimation of Lithium-ion battery plays a key role in battery management system, as it characterizes the battery's health state and ensures safe and stable operation. This paper develops a novel SOH estimation method combining the Gate Recurrent Unit (GRU) and the Beetle Antennae Search Algorithm (BASA). First, the GRU NN with the outstanding capability of the time series and simple structure is optimized by BASA for selecting the optimal hyper-parameters to improve the SOH estimation performance. Second, an efficient features extracted method, called local tangent space alignment (LTSA), is utilized for extracting the crucial features from the measurement data. Finally, several experiments are conducted on a benchmark dataset, and the results show that the proposed method can obtain accurate SOH estimation compared with other methods.