State-of-Health Prediction of lithium-Ion Battery Based on Transfer Learning

作者：Jiayu Hou, Ping Yu, Fengying Zhang, Dingding Liu, Zuxin Li · 年份：2022 · DOI：10.1109/iscsic57216.2022.00047 · 被引用次数：3 · 研究领域：Advanced Battery Technologies Research、Advancements in Battery Materials、Fault Detection and Control Systems

Since the sample data will affect the accuracy of the prediction of the state of health (SOH) of Li-ion batteries based on machine learning, we propose a Transfer learning (TL) method based on Two-stage TrAdaBoost.R2 in this paper. Firstly, the root-mean-square (RMSE) of the source and target domains of the training set is obtained by using support vector regression (SVR), and the parameter α is introduced to adjust the initial weights of these two, and secondly, the similarity between the source domain and the target domain is measured by using the multiple kernel-maximum mean discrepancy (MKMMD), and the weights of the target domain of the training set are updated on this basis. Finally, the proposed algorithm, the original algorithm and the classical machine learning algorithm are compared, and the experimental results show that the RMSE of the proposed method can be less than 0.8% under the condition of insufficient sample data.