Lifetime prognostics of lithium-ion battery pack based on its early cycling data and complete degradation information of battery cells

Lifetime prognostics of lithium-ion batteries plays an important role in improving safety and reducing operation and maintenance costs in the field of energy storage. To rapidly evaluate the lifetime of newly developed battery packs, a method for estimating the future health state of the battery pack using the aging data of the battery cell's full life cycle and the early data of the battery pack is proposed. First, the battery cycle aging characteristics are analyzed from different perspectives. The health indicators (HIs) related to battery aging are extracted from the partial discharge process, and three HIs closely related to battery capacity aging are selected through the Pearson coefficient analysis method. Then, the HIs degradation model of the battery cell based on exponential fitting is corrected by the HIs in the early cycle of the battery pack to predict the HIs degradation curve of each cell in the battery pack in the future cycle. Finally, based on the Gaussian Process Regression (GPR) model, the battery pack's lifetime is predicted using the early 10% cycle data of the battery pack and the predicted HIs of the battery in remaining life cycle. The experimental results show that the mean absolute error (MAE) and root mean squared error (RMSE) of the proposed method are 0.49% and 0.71%, respectively, which verify its advantages of high accuracy and reliability.