Design of State of Health Prediction Model for Retired High Power LiNiMnCoO2 Cell with Holts Exponential Smoothing Method

Abstract

In order to recycle retired battery, it is necessary to know the state of health (SOH) of the retired battery correctly. However, as the battery ages, nonlinearity of the parameter representing SOH becomes more severe. So, the different estimation method is required from the SOH estimation method of a fresh battery. The parameters representing SOH such as discharge capacity, internal resistance, peak point of incremental capacity (IC) curve values that change with aging are time-series data. The SOH estimation of retired batteries requires a technique to analyze nonlinear time-series data. This paper presents design of SOH prediction model for retired high power LiNiMnCoO2 (NCM) cell with holt's exponential smoothing (ES) method. The holt's EX method is the method of nonelinear time-series data analysis. And, the result of the SOH prediction model with the holt's ES is compared with linear regression analysis (LRA) and the moving average (MA) method. © 2020 ACM.