Health State Estimation of Lithium-ion Batteries Based on Discharge Process and Self-Attention-GRU
In order to solve the problem of insufficient feature extraction data for lithium-ion batteries in use and the need for a large amount of historical data in the model,a Self-Attention-GRU based health state estimation method for lithium-ion batteries during discharge process was proposed by analyzing the data of lithium-ion batteries in use.The battery health status was estimated by a model trained on historical data of the same type of lithium battery on lithium batteries without historical data.After having a certain amount of aging data,the aging data of the lithium battery itself was used to train the model and estimate the battery's health status.The equal voltage drop discharge time,root mean square of voltage,and discharge power of the discharge process were extracted as health factors,and the mapping relationship between health factors and state of health(SOH)was established using a gating cycle unit fused with self attention mechanism.Experimental validation was conducted using 4 sets of CALCE battery aging data.When 20%of the aging data was used as the training set,the MAE and RMSE of the model reached 1.03%and 1.25%,respectively.When 30%,40%of the aging data and all aging data of the same type of battery were used as training sets,the MAE and RMSE of the model were both less than or equal to 1%.It was showed that the proposed method had high accuracy and reliability in estimating the health status of lithium-ion bat-teries.
state of healthself-attention mechanismgate recurrent unit(GRU)lithium ion batteryhealth factors
国家科技期刊平台
NSTL
万方数据
