State of Health Estimation for Li-ion Batteries using Improved Gaussian Process Regression and Multiple Health Indicators

Accurately estimating the state of health (SOH) of Li-ion batteries (LIBs) is critical to ensure safe and stable battery operation. However, during the online operation of LIBs, it is difficult to directly measure the SOH. To overcome this challenge, this paper proposes an online SOH estimation method for LIBs based on multiple health factors ( HFs) and improved Gaussian process regression. First, by analysing the dQ/dV curve and dV/dT curve of the LIB, this study finds the data interval with the highest correlation with the battery SOH to extract multidimensional health features. Then, the dimensionality reduction process is performed by the principal component analysis (PCA) method to reduce the computational complexity. Using this approach, a Gaussian process regression (GPR) model with combined kernel functions was proposed to establish the mapping relationship between the predicted values of HFs and SOH. Finally, test experiments are performed on two public lithium battery ageing datasets, and the obtained results show that the estimation error of the proposed method is kept within 1.5%, with high accuracy and reliability.