Abstract This enables the analysis and judgment of the aging degree and consistency of cells within clusters. Meanwhile, neural networks are used to predict the entropy values for short-term health state forecasting of the energy storage station. Finally, the feasibility and effectiveness of the feature data information entropy method for health state assessment and prediction are validated using actual operational data of the energy storage station and a 20S1P battery simulation model. This paper is the first to introduce information entropy theory into the health status assessment of lithium-ion energy storage stations.
Lin et al. (Thu,) studied this question.