A Novel Seasonal Autoregressive Integrated Moving Average Method for the Accurate Lithium-ion Battery Residual Life Prediction

Abstract

Lithium-ion batteries are widely used in electric vehicles (EVs), unmanned aerial vehicles (UAVs), and smart devices because of their high specific energy, long service life, and environmental-friendly. The remaining useful life (RUL) prediction is extremely important for the evaluation of the state of health (SOH) of the battery. Also, it is an important indicator to improve its safety in a variety of applications. In this study, a novel seasonal autoregressive integrated moving average (SARIMA) prediction model is proposed. The proposed model adds periodic parameter optimization to fit the nonlinear characteristics of the battery, including maximum likelihood estimation (MLE) and Akaike information criterion (AIC) to filter the parameters more accurately. So, to effectively solve the shortcomings of the traditional prediction methods, such as complex parameter acquisition, low prediction accuracy, and a large amount of sample data. The method proposed in this paper simplifies the remaining useful life prediction process, ensures high accuracy, and improves the safe operation and reliability of lithium-ion batteries. The model can give the prediction confidence bounds, and the maximum prediction error under complex working conditions is 4.62%.