Estimating state of health of lithium-ion batteries based on generalized regression neural network and quantum genetic algorithm

Abstract

In order to solve the problem of inaccurate estimation of the state of health (SOH) of electric vehicle batteries, this paper proposes a novel SOH estimation algorithm based on particle filter (PF), quantum genetic algorithm (QGA) and generalized regression neural network (GRNN). A denoising method integrating PF and anomaly detection on grouping is proposed to make the network input parameters more stable. To improve estimation accuracy and speed, an optimized GRNN-based SOH estimation model is proposed. Based on the advantages of GRNN with fewer layers and fewer hyperparameters, the Pearson correlation coefficient and QGA are used to optimize its weights to realize the adaptive determination of hyperparameters. The experiment results based on NASA and the real vehicle dataset show that the proposed algorithm has the advantages of high estimation accuracy and low computational cost, which is of great significance to the SOH estimation of electric vehicle batteries under actual operating conditions.