State-of-Health Prediction Using Transfer Learning and a Multi-Feature Fusion Model

Pengyu Fu; Liang Chu; Zhuoran Hou; Zhiqi Guo; Yang Lin; Jincheng Hu

doi:10.3390/s22218530

State-of-Health Prediction Using Transfer Learning and a Multi-Feature Fusion Model

Sensors (Basel). 2022 Nov 5;22(21):8530. doi: 10.3390/s22218530.

Authors

Pengyu Fu¹, Liang Chu¹, Zhuoran Hou¹, Zhiqi Guo¹, Yang Lin¹, Jincheng Hu²

Affiliations

¹ College of Automotive Engineering, Jilin University, Changchun 130022, China.
² Department of Aeronautical and Automotive Engineering, Loughborough University, Loughborough LE11 3TU, UK.

Abstract

Existing data-driven technology for prediction of state of health (SOH) has insufficient feature extraction capability and limited application scope. To deal with this challenge, this paper proposes a battery SOH prediction model based on multi-feature fusion. The model is based on a convolutional neural network (CNN) and a long short-term memory network (LSTM). The CNN can learn the cycle features in the battery data, the LSTM can learn the aging features of the battery over time, and regression prediction can be made through the full-connection layer (FC). In addition, for the aging differences caused by different battery operating conditions, this paper introduces transfer learning (TL) to improve the prediction effect. Across cycle data of the same battery under 12 different charging conditions, the fusion model in this paper shows higher prediction accuracy than with either LSTM and CNN in isolation, reducing RMSPE by 0.21% and 0.19%, respectively.

Keywords: convolutional neural network (CNN); long short-term memory (LSTM); multi-feature fusion (MFF); state of health (SOH); transfer learning (TL).

MeSH terms

Machine Learning*
Neural Networks, Computer*

Grants and funding

This research received no external funding.