【摘要】 锂电池剩余使用寿命（Remaining useful life,RUL）预测是锂电池研究的一个重要方向，通过对RUL的准确预测，可以更好地管理和维护电池，延长电池使用寿命。为了能够准确预测锂电池的RUL，提出了一种集合变分模态分解（Variational mode decomposition,VMD）、长短时记忆网络（Long short-term memory,LSTM）和自回归移动平均模型（Autoregressive integrated moving average,ARIMA）相结合的锂电池RUL预测模型。该模型首先采用VMD算法将NASA锂电池数据集中的容量数据分解为多个高频分量和低频分量，以此减少容量数据中的噪声干扰，然后针对各个分量的特点，分别利用LSTM和ARIMA对分解所得的高频分量和低频分量建立预测子模型，最后将各个子模型的预测值进行叠加重构得到锂电池的RUL结果。实验结果表明VMD-LSTM-ARIMA预测模型相比于其他预测模型，该模型具有较好的锂电池RUL预测能力。并在CALCE锂电池数据集上进行了泛化性实验，结果表明该模型适用于不同电池RUL预测任务。更多还原

【Abstract】 The prediction of remaining useful life(RUL) of lithium-ion batteries is an important research direction in battery technology.Through accurate prediction of RUL,batteries can be better managed and maintained to extend their lifespan.To achieve accurate RUL prediction of lithium-ion batteries,a model which combines variational mode decomposition(VMD) with long short-term memory(LSTM) and autoregressive integrated moving average(ARIMA) was proposed.Firstly,VMD algorithm was used to decompose the capacity data from the NASA lithium-ion battery dataset into multiple high-frequency and low-frequency components in order to reduce the noise interference in the capacity data.Then,with regard to the characteristics of each component,LSTM and ARIMA were used to establish separate sub-models to predict the high-frequency and low-frequency components,respectively.Finally,the predicted values of each sub-model were combined and reconstructed to obtain the RUL result of the lithium-ion battery.Experimental results showed that the VMDLSTM-ARIMA prediction model had better RUL prediction capability compared with other prediction models.Furthermore,generalization experiments on the CALCE lithium-ion battery dataset showed that the model was applicable to different battery RUL prediction tasks.更多还原