基于改进粒子滤波算法实现锂离子电池RUL预测

RUL Prediction of Lithium-ion Battery Based on Improved Particle Filtering Algorithm

鉴于采用传统粒子滤波算法来预测锂离子电池剩余使用寿命(RUL)过程中,存在粒子多样性丧失现象而导致RUL预测精度较低的问题,引入线性优化重采样思想,建立了基于线性优化重采样粒子滤波(LORPF)的电池RUL预测方法。该方法以双指数模型作为电池老化模型,通过LORPF算法对模型参数进行迭代更新,实现电池RUL预测并给出预测结果的不确定性表达,最后使用美国国家航空航天局PCoE研究中心的电池数据和自主搭建实验平台的电池数据对所提方法与传统PF方法进行对比验证,结果表明该方法有效提高了RUL预测精度,其误差小于5%。

When using traditional particle filtering(PF)algorithm to predict the remaining useful life(RUL)of lithium-ion battery,a phenomenon of losing particle diversity occurs,leading to low accuracy in RUL prediction.In view of this,an idea of linear optimization resampling is introduced to establish a battery RUL prediction method based on linear optimization resampling particle filtering(LORPF).The method adopts the double exponential model as battery aging model,and the model parameters are iterated and updated by LORPF algorithm to achieve battery RUL prediction with an uncertainty expression of prediction result given.Finally,the method proposed and the traditional PF method are comparatively verified based on the battery data of NASA PCoE in the US and the experimental platform self-built.The results show that the method proposed effectively enhances the RUL prediction accuracy with an error less than 5%.