An improved fractional‐order state estimation algorithm based on an unscented particle filter for state of charge estimation of lithium‐ion batteries with adaptive estimations of unknown parameters

Zhiyuan Jiao,Zhe Gao,Haoyu Chai,Dandan Song
DOI: https://doi.org/10.1002/cta.3862
IF: 2.378
2023-11-29
International Journal of Circuit Theory and Applications
Abstract:An adaptive fractional‐order unscented particle filter with the joint estimations of order and unknown parameters is proposed to estimate the state of charge (SOC) of lithium‐ion batteries. Besides, the adaptive estimations of the noise covariance matrices are applied to improve the accuracy of SOC estimation of the proposed algorithm. Summary An adaptive fractional‐order unscented particle filter (FOUPF) and an adaptive FOUPF with the estimations of the noise covariance matrices (NCMs) are proposed to estimate the state of charge (SOC) of lithium‐ion batteries (LIBs) in this study. The Grünwald–Letnikov difference is employed to establish the corresponding discrete‐time equation of the fractional‐order system (FOS). The order and SOC of FOS of LIBs are controlled within an appropriate interval by a mapping function. In order to achieve a joint estimation of order and unknown parameters for a FOS, an augmented vector method is applied in this study. Compared with the adaptive fractional‐order unscented Kalman filter and adaptive fractional‐order cubature Kalman filter, the proposed adaptive FOUPF has a higher accuracy for estimating SOC. Besides, an iterative approach that accommodates the NCMs is proposed to improve the estimation accuracy of SOC. Finally, the availability of the proposed algorithms is tested by several experiments.
engineering, electrical & electronic
What problem does this paper attempt to address?