DGM和RVM融合的电动汽车电池寿命预测研究

Research on Battery Life Prediction of Electric Vehicle Based on DGM and RVM

电池寿命预测已经成为指导电池安全运行维护的一个重要依据,针对相关向量机(relevance vector machine,RVM)和灰色模型(grey model,GM)等传统方法中存在的长期预测能力差和预测精度低、稳定性不高等问题,提出了一种离散灰色模型(diecrete grey model,DGM)和RVM融合的电池寿命预测方法,该方法利用DGM的动态更新提高了RVM的预测精度,从而解决了电动汽车预测精度不高和稳定性差的问题,实验结果证明,电动汽车电池寿命预测能力得到了极大的提高,其预测精度可达95%以上,完全满足电动汽车电池运行维护的要求。

Battery life prediction has become an important basis for guiding the safe operation and maintenance of the battery. The long-term predictive ability of the relevance vector machine is poor. The prediction accuracy and stability of gray model is low. In response to these problems, a battery life prediction method based on discrete gray model and RVM is proposed in this paper. This method improves the prediction accuracy of RVM by dynamic updating DGM. It solves the problems of low prediction accuracy and low stability for electric vehicles. The experimental results show that the battery life prediction ability of electric vehicle has been greatly improved and the prediction accuracy up to 95%. It fully meets the requirements of operation and maintenance of battery for electric vehicle.