Multiple-imputation-particle-filtering scheme for Uncertainty Characterization in Battery State-of-Charge Estimation Problems with Missing Measurement Data

Abstract

The design of particle-filtering-based algorithms for estimation often has to deal with the problem of missing observations. This requires the implementation of an appropriate methodology for real-time uncertainty characterization, within the estimation process, incorporating knowledge from other available sources of information. This article presents preliminary results of a multiple imputation strategy used to improve the performance of a particle-filtering-based state- of-charge (SOC) estimator for lithium-ion (Li-Ion) battery cells. The proposed uncertainty characterization scheme is tested and validated in a case study where the state-space model requires both voltage and discharge current measurements to estimate the SOC. A sudden disconnection of the battery’s voltage sensor is assumed to cause significant loss of data. The results show that the multiple-imputation particle filter enables reasonable uncertainty characterization for the state estimate as long as the voltage sensor disconnection continues. Furthermore, when the voltage measurements are once more available, the level of uncertainty adjusts to levels that are comparable to the case where data was not lost.