The state of health prediction of Li-ion batteries based on an improved extreme learning machine

In this paper, in order to accurately predict the state of health (SOH) of lithium-ion (Li-ion) batteries in real time and ensure the safe operation of any related equipment, health factors that can characterize battery degradation were extracted from charging data, and the correlations between health factors and battery capacity were analyzed using the Spearman and Pearson coefficients. Furthermore, an extreme learning machine (ELM) prediction method that was optimized based on the Beetle Antennae Search (BAS) algorithm was proposed for the online prediction of the SOH of Li-ion batteries, and finally, the proposed model was validated using the NASA battery dataset. The results indicate that the proposed BAS-ELM method can predict the SOH of Li-ion batteries more accurately than the ELM and back propagation methods.