Investigation of a data-driven SOC estimator based on the merged SMO and degradation mitigation for series/parallel-cell configured battery pack

Because of sequential cycling of series/parallel-cell configured battery pack, cell´s degradation in the battery pack is absolutely inevitable. Then, in order to prevent an additional degradation of the pack, an accurate knowledge of state-of-charge (SOC) estimator for deteriorated pack is essential. Thus, in this approach, a data-driven SOC estimation method based on the merged sliding-mode observer (SMO) and degradation mitigation considering recursive least squares (RLS) for aged battery pack is newly introduced. Specifically, for application of the proposed work for series/parallel-cell configured battery pack, discrimination process that determines cells with similar electrochemical characteristics was previously implemented. In addition, battery pack modeling based on the discrimination process was applied to the SMO for improved SOC estimation of degraded battery pack. For validation of the proposed work, A degraded Li-Ion cell of 2.2Ah produced by LG Chem was used to construct experimental series/parallel-cell configured battery packs of 4S1P and 3S2P.