State-of-health Estimation of Lithium-ion Batteries Based on EMD-DO-Elman and GRA
Accurate and reliable state-of-health(SOH)estimation of lithium-ion batteries can help improve battery equipment's safety and stability.This paper proposes an EMD-DO-Elman method for estimating the SOH of lithium-ion batteries in response to current issues such as the inability to measure SOH directly,difficulty extracting health features,and insufficient estimation methods.Based on the lithium-ion battery aging test data publicly available at NASA Ames Research Center and actual experimental test battery data,it is proposed to use empirical mode decomposition(EMD)to decompose the battery aging data into signal decomposition,to obtain the characteristic components reflecting the battery SOH.Besides,grey relation analysis(GRA)is used to conduct correlation analysis on the characteristic components to select model inputs.Finally,the dandelion optimizer(DO)is applied to optimize the parameters of the Elman network to improve the estimation performance.The experimental results show that this method can accurately estimate the SOH of lithium-ion batteries,and the R2 of the estimation results is always greater than 98％.In addition,verifying the SOH estimation of battery data under different training set quantities further proves that the estimation model proposed in this paper has good generalization and robustness.
万方数据
维普
