A Virtual Sample Generation Method Based on Differential Evolution Algorithm for Overall Trend of Small Sample Data: Used for Lithium-ion Battery Capacity Degradation Data

Guoqing Kang,Lifeng Wu,Zhen Peng,Yong Guan

doi:10.1109/access.2019.2937550

Abstract

Considering the wide application of lithium-ion battery in life, the prediction of the remaining life of lithium-ion battery has become a research hotspot. Studies show, due to the improvement of the technology level of lithium-ion battery, its life is getting longer and longer. Even under the condition of accelerated life test, it is difficult to obtain enough available data for research in a short term. In order to solve the problem of how to accurately predict the residual life with the data-driven method under the condition of small sample size, an overall trend virtual sample generation method based on differential evolution (OT-DEVSG) is proposed. This method uses a differential evolution algorithm with better optimization performance, and improves the original mega-trend-diffusion (MTD) method, the range of virtual samples is effectively constrained and the trend of samples can be estimated more accurately. The method can effectively generate a virtual sample data sequence with time parameters, and adapt the virtual sample to the real-life sample trend, which solves the problem of insufficient degradation data of the lithium-ion batteries. Finally, we validate the effectiveness of the OT-DEVSG method with three existing data sets. The experimental results show that the proposed OT-DEVSG method is effective for solving the problem of long-term life prediction of lithium-ion batteries.

Highlights

Lithium-ion battery is an environmentally-friendly highenergy rechargeable battery
The OT-DEVSG method given in this paper can generate a virtual sample sequence effectively, It solves the existing small sample problem very well and improve the predicting accuracy of Back Propagation Neural Networks (BPNN) which achieves the accurate long-term prediction of the remaining life of the lithium-ion battery at any time point
PROPOSED METHOD In this paper, a method of generating overall trend virtual samples based on differential evolution algorithm is proposed for small samples, and BPNN is established to test the reliability of virtual samples

Summary

INTRODUCTION

Lithium-ion battery is an environmentally-friendly highenergy rechargeable battery. Because of its various advantages, such as high capacity, low self-discharge rate, high safety and long cycle life, it is widely used in areas like electronic communication engineering, transportation and aerospace [1], [2]. Chen et al [16] proposed a virtual sample generation method based on PSO When it comes to analyzing the degradation data of lithium-ion battery with time attribute, these methods are not applicable. They cannot describe the overall trend of data well or solve the problem of inaccurate prediction of the remaining life of lithium-ion battery in the long term. The OT-DEVSG method given in this paper can generate a virtual sample sequence effectively, It solves the existing small sample problem very well and improve the predicting accuracy of Back Propagation Neural Networks (BPNN) which achieves the accurate long-term prediction of the remaining life of the lithium-ion battery at any time point. The validity of the proposed method is verified by a comparison between it and the PSOVSG method as well as experiments under three different data sets

PROPOSED METHOD

Findings

CONCLUSION