State of charge estimation for lithium-ion batteries based on a novel complex-order model

• Tipo: Artigo em Revista Científica Internacional
• Data: 2023

Autores

Chen, LP (Autor) Outra; Wu, XB (Autor) Outra; António Mendes Lopes (Autor) FEUP; Li, X (Autor) Outra; Li, PH (Autor) Outra; Wu, RC (Autor) Outra

Revista

Título: Communications in Nonlinear Science and Numerical Simulation

Vol. 125

ISSN: 1007-5704

Editora: Elsevier

Indexação

ISI Web of Knowledge - 0 Citações

Scopus

Outras Informações

ID Authenticus: P-00Y-HM3

DOI: 10.1016/j.cnsns.2023.107365

Abstract (EN): The accuracy of the battery model is decisive in model-based state of charge (SOC) estimation. In this paper, complex-order derivatives (CDs) are applied in the scope of battery modeling, parameter identification, and SOC estimation. Firstly, a novel complexorder equivalent circuit model (Co-ECM) for lithium-ion batteries, which considers an innovative complex-order constant phase element, is proposed. Secondly, the structure characteristics of the Co-ECM are analyzed, and a complex-order particle swarm optimization algorithm is developed to identify the Co-ECM parameters. Finally, a novel complex-order unscented Kalman filter is designed to estimate the battery SOC, while CDs capture the system past behavior and tackle the nonlinearities of the constant phase element. Also, the proposed Co-ECM is compared with two other alternatives (i.e., integer-order and fractional-order ECM) based on data from two battery test cycles at different temperatures. The results show that the new Co-ECM leads to SOC estimation accuracy higher than the traditional models over a wide range of temperature (0 degrees C, 25 degrees C and 45 degrees C), with root-mean-squared error (RMSE) and mean absolute error (MAE) less than 0.47% and 0.41%, respectively. Moreover, experiments with data polluted with artificial noise revealed that the proposed model has superior robustness against noisy information. The new Co-ECM is, thus, shown to be a prime option for battery SOC estimation.& COPY; 2023 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 19