Improved lithium battery state of health estimation and enhanced adaptive capacity of innovative kernel extreme learning machine optimized by multi-strategy dung beetle algorithm.

Mo, Daijiang; Wang, Shunli; Zhang, Mengyun; Fan, Yongcun; Wu, Wenjie; Fernandez, Carlos; Su, Qiyong

Abstract

Accurate estimation of the state of health (SOH) of lithium batteries is crucial to ensure the reliable and safe operation of lithium batteries. Aiming at the problems of low accuracy of extreme learning machine and poor mapping ability of conventional kernel function, this paper constructs a kernel extreme learning machine model and uses a multi-strategy improved dung beetle algorithm to find the optimal parameters. In this paper, for the poor estimation effect caused by the difficulty of adapting the conventional kernel function to nonlinear batteries, we design a cosine polynomial kernel function, which improves the linear divisibility of the data; in addition, for the global search, local development, and convergence improvement of the dung beetle algorithm, we introduce the optimal Latin hypercubic idea, the Cauchy variation strategy, and the sparrow alert mechanism, which successfully improve the parameter searching capability and sensitivity of the algorithm, respectively. We successfully improve the capability and sensitivity of the algorithm in parameter searching. We experimentally verify the reliability and validity of the proposed model, and the maximum root mean square error and the average absolute percentage error obtained in the test are not higher than 0.00753 and 0.00399, respectively, and the minimum fit is not lower than 0.9921, which reflects the high accuracy and strong adaptive ability of the model.

Citation

MO, D., WANG, S., ZHANG, M., FAN, Y., WU, W., FERNANDEZ, C. and SU, Q. [2024]. Improved lithium battery state of health estimation and enhanced adaptive capacity of innovative kernel extreme learning machine optimized by multi-strategy dung beetle algorithm. Ionics [online], Latest Articles. Available from: https://doi.org/10.1007/s11581-024-05914-6

• Journal Article Type: Article
• Acceptance Date: Oct 27, 2024
• Online Publication Date: Nov 18, 2024
• Deposit Date: Dec 3, 2024
• Publicly Available Date: Nov 19, 2025
• Journal: Ionics
• Print ISSN: 0947-7047
• Electronic ISSN: 1862-0760
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• DOI: https://doi.org/10.1007/s11581-024-05914-6
• Keywords: Cosine polynomial kernel function; Dung beetle optimization algorithm; Kernel extreme learning machine; Lithium battery state of health; Sparrow algorithm
• Public URL: https://rgu-repository.worktribe.com/output/2593207