A comparative study of global optimization methods for parameter identification of different equivalent circuit models for Li-ion batteries. (1st February 2019)

• Record Type: Journal Article
• Authors: Lai, Xin; Gao, Wenkai; Zheng, Yuejiu; Ouyang, Minggao; Li, Jianqiu; Han, Xuebing; Zhou, Long
• Is Part Of: Electrochimica acta. Volume 295(2019)
• Page Start: 1057
• Page End: 1066
• Publication Date: 2019-02-01
• Subjects: Equivalent circuit model -- Parameter identification -- Optimization algorithm -- Metaheuristic algorithm -- Li-ion battery; Electrochemistry -- Periodicals; Electrochemistry, Industrial -- Periodicals; 541.37
• DOI: 10.1016/j.electacta.2018.11.134
• Languages: English
• ISSNs: 0013-4686

Abstract

Abstract: A suitable model structure and matched model parameters are prerequisites for the precise estimation of the battery states. Previous studies pay little attention to whether a parameter identification method is suitable for a model. In this study, a comparative study is conducted by implementing model parameter optimization for nine equivalent circuit models using nine optimizers in the entire SOC area. The following conclusions are drawn: (1) PNGV and the exact algorithms are an ideal combination in the low SOC area (0%–20%). (2) In the high SOC area (20–100%), exact algorithms are an ideal choice for the first-order RC models, and PSO is an ideal identification algorithm for second-order RC models. For the third- and fourth-order RC models, firefly algorithm has the highest accuracy with longer identification time. (3) Firefly algorithm has the superior capacity to identify the accurate model parameters and PSO has the best comprehensive performance for on-line parameter identification. Highlights: A comparative study on model parameter identification for nine models using nine optimizers in the entire SOC area. PNGV and exact algorithms are an ideal combination in the low SOC area. In the high SOC area, EAs and PSO are the ideal choice for the first- and second-order RC models, respectively. For the third- and fourth-order RC models, firefly algorithm has the highest accuracy with longer identification time. FA has excellent identification accuracy and PSO has the best comprehensive performance for on-line identification.