Research on Lithium-Ion Battery SOC Estimation Based on Ultrasonic Sensing and Gaussian Process Regression

The use of ultrasound to estimate the state of charge（SOC） of lithium-ion batteries has been initially verified. Aiming at the problems of small sample size, single experimental conditions, and imperfect model establishment in the existing ultrasonic SOC estimation, a lithium battery SOC estimation method based on cross-validation Gaussian process regression is proposed. Experiment with lithium batteries at different temperatures and different working conditions, and conduct a preliminary analysis of the data obtained; then analyze the data at a single temperature, and establish a cross-validation Gaussian process regression model based on the joint amplitude transit time（PT-RPGPR）; For the data obtained from experiments at different temperatures, a new model（T-PT-RPGPR） is built by adding temperature factors. The experimental results show that combining ultrasonic information and Gaussian process to build a model for lithium battery SOC estimation under complex working conditions has high estimation accuracy.