基于数据驱动的卫星锂离子电池寿命预测方法

Research on Data-driven Life Prediction Methods of Satellite Lithium-ion Battery

锂离子电池由于具有工作电压高、质量轻、比能量高、寿命长和自放电率小等优点,成为替代传统镍氢、镍镉电池的第3代航天器用储能电源;寿命预测是锂离子电池健康管理的重要方面,是掌握电源性能衰退趋势的重要手段,锂离子电池寿命预测问题已成为电子系统健康管理领域的研究热点;针对锂离子电池的寿命预测问题,采用了NASA埃姆斯中心的锂离子电池地面试验采集的数据,然后重点研究了3种基于数据驱动的方法,并对锂离子电池的寿命进行了估计,最后对各种预测方法的效果进行了评价;实验结果表明,文本提出的基于数据驱动的方法能够有效地用于锂离子电池寿命预测中,在工程应用方面具有较高的实际价值。

Lithium ion battery has many advantages, such as high working voltage, light weight, high specific energy, long life and small self --discharge rate. It＇ s a alternative to traditional Ni--Mh battery or Ni--Ca battery and which becomes the third generation power for space energy storage. Life prediction is an important aspect in health management of Lithium--ion battery, it is also an important mean to understand the power performance decline. The reasearch on Life prediction method of Lithium--ion battery becomes a research hotspot in electronic health management system. Aiming at the life prediction methods of Lithium--ion battery, we adopted the experiment data which comes from the ground test sets of the NASA Ames Center. Then we studied three kinds of data--driven prediction methods, and estimated the life of Lithium ion battery. At last, we an alysed the effects of various prediction methods. Experimental results show that the proposed data--driven prediction method can be effectively used in life prediction of Lithium--ion battery, and it has strong practical value in engnneering application.