Remaining useful life estimation of lithium-ion battery using exemplar-based conditional particle filter

Since lithium-ion batteries have been used in a wide range of fields, such as transportation industry, household appliances, and national defence industry. In order to avoid the unnecessary loss resulting from its sudden failure, it is necessary to timely predict the remaining useful life (RUL) of lithium-ion battery. In this paper, we present a novel remaining useful life estimation method for lithium-ion batteries, which depends on exemplar-based conditional particle filter (EC-PF). Differently from traditional particle filter, in the update phase, exemplar-based conditional particle filter combines historical data of multiple batteries with filtering stage of a single battery to compute the weights with respect to particles. This method can make the weights of particles more accurate, which results in improving the prediction accuracy. To verify the effectiveness and efficiency of the proposed method, a public data set is selected for validating prediction accuracy of RUL of battery. The results show that the proposed method improves the performance of the traditional particle filter method.