Prediction of Health Level of Multiform Lithium Sulfur Batteries Based on Incremental Capacity Analysis and an Improved LSTM

• DOI: 10.23919/pcmp.2023.000280
• Publication Date: 2024-03-07T19:13:53Z
• Authors: Hao Zhang; Hanlei Sun; Le Kang; Yi Zhang; Licheng Wang; Kai Wang

Capacity estimation plays a crucial role in battery management systems, and is essential for ensuring the safety reliability of lithium-sulfur (Li-S) batteries. This paper proposes method that uses long short-term memory (LSTM) neural network to estimate state health (SOH) Li-S The features extracted from charging curve incremental capacity analysis (ICA) as input LSTM network. To enhance robustness accuracy network, Adam algorithm employed optimize specific hyperparameters. Experimental data three different groups batteries with varying nominal capacities are used validate proposed method. results demonstrate effectiveness accurately estimating degradation all Also, study examines impact lengths training sets on estimation. reveal ICA-LSTM model achieves prediction mean absolute error 4.6% squared 0.21% set 20%, 40%, 60%. demonstrates lightweight maintains high SOH even small set, exhibits strong adaptive generalization capabilities when applied Overall, method, supported by experimental validation analysis, its efficacy accurate reliable estimation, thereby enhancing performance